Upload folder using huggingface_hub
Browse files- added_tokens.json +40 -0
- config.json +30 -0
- generation_config.json +4 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +327 -0
- trainer_state.json +3533 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"\t\t": 50294,
|
| 3 |
+
"\t\t\t": 50293,
|
| 4 |
+
"\t\t\t\t": 50292,
|
| 5 |
+
"\t\t\t\t\t": 50291,
|
| 6 |
+
"\t\t\t\t\t\t": 50290,
|
| 7 |
+
"\t\t\t\t\t\t\t": 50289,
|
| 8 |
+
"\t\t\t\t\t\t\t\t": 50288,
|
| 9 |
+
"\t\t\t\t\t\t\t\t\t": 50287,
|
| 10 |
+
" ": 50286,
|
| 11 |
+
" ": 50285,
|
| 12 |
+
" ": 50284,
|
| 13 |
+
" ": 50283,
|
| 14 |
+
" ": 50282,
|
| 15 |
+
" ": 50281,
|
| 16 |
+
" ": 50280,
|
| 17 |
+
" ": 50279,
|
| 18 |
+
" ": 50278,
|
| 19 |
+
" ": 50277,
|
| 20 |
+
" ": 50276,
|
| 21 |
+
" ": 50275,
|
| 22 |
+
" ": 50274,
|
| 23 |
+
" ": 50273,
|
| 24 |
+
" ": 50272,
|
| 25 |
+
" ": 50271,
|
| 26 |
+
" ": 50270,
|
| 27 |
+
" ": 50269,
|
| 28 |
+
" ": 50268,
|
| 29 |
+
" ": 50267,
|
| 30 |
+
" ": 50266,
|
| 31 |
+
" ": 50265,
|
| 32 |
+
" ": 50264,
|
| 33 |
+
" ": 50263,
|
| 34 |
+
" ": 50262,
|
| 35 |
+
" ": 50261,
|
| 36 |
+
" ": 50260,
|
| 37 |
+
" ": 50259,
|
| 38 |
+
" ": 50258,
|
| 39 |
+
" ": 50257
|
| 40 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "semran1/56_7_62",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"QPhiForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"eos_token_id": null,
|
| 10 |
+
"hidden_act": "gelu_new",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 8192,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"model_type": "phi",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 24,
|
| 19 |
+
"num_key_value_heads": 32,
|
| 20 |
+
"partial_rotary_factor": 0.5,
|
| 21 |
+
"qk_layernorm": false,
|
| 22 |
+
"resid_pdrop": 0.0,
|
| 23 |
+
"rope_scaling": null,
|
| 24 |
+
"rope_theta": 10000.0,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"torch_dtype": "bfloat16",
|
| 27 |
+
"transformers_version": "4.48.3",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 51200
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.48.3"
|
| 4 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5835a6d88e23f0e35002825ecd629545814789c892b5f6aa4b8aa3f1be7f069d
|
| 3 |
+
size 2837479464
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:775994ef16e0016c3c5dd55b947ff3b284c90c44e1bc0ba23062ff731507d323
|
| 3 |
+
size 2672712800
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4c917636c7a58af68a29056522a757e9f9b99005b776641aa157c536967817d
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:215cf579801eeae11f5d1f5aad10ba64e61bc8600878bbf41515b4e3d66e79ae
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,327 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"50257": {
|
| 13 |
+
"content": " ",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": false
|
| 19 |
+
},
|
| 20 |
+
"50258": {
|
| 21 |
+
"content": " ",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": false
|
| 27 |
+
},
|
| 28 |
+
"50259": {
|
| 29 |
+
"content": " ",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": false
|
| 35 |
+
},
|
| 36 |
+
"50260": {
|
| 37 |
+
"content": " ",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": false
|
| 43 |
+
},
|
| 44 |
+
"50261": {
|
| 45 |
+
"content": " ",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": false
|
| 51 |
+
},
|
| 52 |
+
"50262": {
|
| 53 |
+
"content": " ",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": false
|
| 59 |
+
},
|
| 60 |
+
"50263": {
|
| 61 |
+
"content": " ",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": true,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": false
|
| 67 |
+
},
|
| 68 |
+
"50264": {
|
| 69 |
+
"content": " ",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": true,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": false
|
| 75 |
+
},
|
| 76 |
+
"50265": {
|
| 77 |
+
"content": " ",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": true,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": false
|
| 83 |
+
},
|
| 84 |
+
"50266": {
|
| 85 |
+
"content": " ",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": true,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": false
|
| 91 |
+
},
|
| 92 |
+
"50267": {
|
| 93 |
+
"content": " ",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": true,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": false
|
| 99 |
+
},
|
| 100 |
+
"50268": {
|
| 101 |
+
"content": " ",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": true,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": false
|
| 107 |
+
},
|
| 108 |
+
"50269": {
|
| 109 |
+
"content": " ",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": true,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": false
|
| 115 |
+
},
|
| 116 |
+
"50270": {
|
| 117 |
+
"content": " ",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": true,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": false
|
| 123 |
+
},
|
| 124 |
+
"50271": {
|
| 125 |
+
"content": " ",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": true,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": false
|
| 131 |
+
},
|
| 132 |
+
"50272": {
|
| 133 |
+
"content": " ",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": true,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": false
|
| 139 |
+
},
|
| 140 |
+
"50273": {
|
| 141 |
+
"content": " ",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": true,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": false
|
| 147 |
+
},
|
| 148 |
+
"50274": {
|
| 149 |
+
"content": " ",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": true,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": false
|
| 155 |
+
},
|
| 156 |
+
"50275": {
|
| 157 |
+
"content": " ",
|
| 158 |
+
"lstrip": false,
|
| 159 |
+
"normalized": true,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": false
|
| 163 |
+
},
|
| 164 |
+
"50276": {
|
| 165 |
+
"content": " ",
|
| 166 |
+
"lstrip": false,
|
| 167 |
+
"normalized": true,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": false
|
| 171 |
+
},
|
| 172 |
+
"50277": {
|
| 173 |
+
"content": " ",
|
| 174 |
+
"lstrip": false,
|
| 175 |
+
"normalized": true,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": false
|
| 179 |
+
},
|
| 180 |
+
"50278": {
|
| 181 |
+
"content": " ",
|
| 182 |
+
"lstrip": false,
|
| 183 |
+
"normalized": true,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": false
|
| 187 |
+
},
|
| 188 |
+
"50279": {
|
| 189 |
+
"content": " ",
|
| 190 |
+
"lstrip": false,
|
| 191 |
+
"normalized": true,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": false
|
| 195 |
+
},
|
| 196 |
+
"50280": {
|
| 197 |
+
"content": " ",
|
| 198 |
+
"lstrip": false,
|
| 199 |
+
"normalized": true,
|
| 200 |
+
"rstrip": false,
|
| 201 |
+
"single_word": false,
|
| 202 |
+
"special": false
|
| 203 |
+
},
|
| 204 |
+
"50281": {
|
| 205 |
+
"content": " ",
|
| 206 |
+
"lstrip": false,
|
| 207 |
+
"normalized": true,
|
| 208 |
+
"rstrip": false,
|
| 209 |
+
"single_word": false,
|
| 210 |
+
"special": false
|
| 211 |
+
},
|
| 212 |
+
"50282": {
|
| 213 |
+
"content": " ",
|
| 214 |
+
"lstrip": false,
|
| 215 |
+
"normalized": true,
|
| 216 |
+
"rstrip": false,
|
| 217 |
+
"single_word": false,
|
| 218 |
+
"special": false
|
| 219 |
+
},
|
| 220 |
+
"50283": {
|
| 221 |
+
"content": " ",
|
| 222 |
+
"lstrip": false,
|
| 223 |
+
"normalized": true,
|
| 224 |
+
"rstrip": false,
|
| 225 |
+
"single_word": false,
|
| 226 |
+
"special": false
|
| 227 |
+
},
|
| 228 |
+
"50284": {
|
| 229 |
+
"content": " ",
|
| 230 |
+
"lstrip": false,
|
| 231 |
+
"normalized": true,
|
| 232 |
+
"rstrip": false,
|
| 233 |
+
"single_word": false,
|
| 234 |
+
"special": false
|
| 235 |
+
},
|
| 236 |
+
"50285": {
|
| 237 |
+
"content": " ",
|
| 238 |
+
"lstrip": false,
|
| 239 |
+
"normalized": true,
|
| 240 |
+
"rstrip": false,
|
| 241 |
+
"single_word": false,
|
| 242 |
+
"special": false
|
| 243 |
+
},
|
| 244 |
+
"50286": {
|
| 245 |
+
"content": " ",
|
| 246 |
+
"lstrip": false,
|
| 247 |
+
"normalized": true,
|
| 248 |
+
"rstrip": false,
|
| 249 |
+
"single_word": false,
|
| 250 |
+
"special": false
|
| 251 |
+
},
|
| 252 |
+
"50287": {
|
| 253 |
+
"content": "\t\t\t\t\t\t\t\t\t",
|
| 254 |
+
"lstrip": false,
|
| 255 |
+
"normalized": true,
|
| 256 |
+
"rstrip": false,
|
| 257 |
+
"single_word": false,
|
| 258 |
+
"special": false
|
| 259 |
+
},
|
| 260 |
+
"50288": {
|
| 261 |
+
"content": "\t\t\t\t\t\t\t\t",
|
| 262 |
+
"lstrip": false,
|
| 263 |
+
"normalized": true,
|
| 264 |
+
"rstrip": false,
|
| 265 |
+
"single_word": false,
|
| 266 |
+
"special": false
|
| 267 |
+
},
|
| 268 |
+
"50289": {
|
| 269 |
+
"content": "\t\t\t\t\t\t\t",
|
| 270 |
+
"lstrip": false,
|
| 271 |
+
"normalized": true,
|
| 272 |
+
"rstrip": false,
|
| 273 |
+
"single_word": false,
|
| 274 |
+
"special": false
|
| 275 |
+
},
|
| 276 |
+
"50290": {
|
| 277 |
+
"content": "\t\t\t\t\t\t",
|
| 278 |
+
"lstrip": false,
|
| 279 |
+
"normalized": true,
|
| 280 |
+
"rstrip": false,
|
| 281 |
+
"single_word": false,
|
| 282 |
+
"special": false
|
| 283 |
+
},
|
| 284 |
+
"50291": {
|
| 285 |
+
"content": "\t\t\t\t\t",
|
| 286 |
+
"lstrip": false,
|
| 287 |
+
"normalized": true,
|
| 288 |
+
"rstrip": false,
|
| 289 |
+
"single_word": false,
|
| 290 |
+
"special": false
|
| 291 |
+
},
|
| 292 |
+
"50292": {
|
| 293 |
+
"content": "\t\t\t\t",
|
| 294 |
+
"lstrip": false,
|
| 295 |
+
"normalized": true,
|
| 296 |
+
"rstrip": false,
|
| 297 |
+
"single_word": false,
|
| 298 |
+
"special": false
|
| 299 |
+
},
|
| 300 |
+
"50293": {
|
| 301 |
+
"content": "\t\t\t",
|
| 302 |
+
"lstrip": false,
|
| 303 |
+
"normalized": true,
|
| 304 |
+
"rstrip": false,
|
| 305 |
+
"single_word": false,
|
| 306 |
+
"special": false
|
| 307 |
+
},
|
| 308 |
+
"50294": {
|
| 309 |
+
"content": "\t\t",
|
| 310 |
+
"lstrip": false,
|
| 311 |
+
"normalized": true,
|
| 312 |
+
"rstrip": false,
|
| 313 |
+
"single_word": false,
|
| 314 |
+
"special": false
|
| 315 |
+
}
|
| 316 |
+
},
|
| 317 |
+
"bos_token": "<|endoftext|>",
|
| 318 |
+
"chat_template": "\n{%- for message in messages %}\n {%- if message['from'] == 'human' %}\n {{- 'user' ~ '\n' ~ message['value'] ~ '<|endoftext|>' }}\n {%- elif message['from'] == 'gpt' %}\n {{- 'assistant' ~ '\n' ~ message['value'] ~ '<|endoftext|>' }}\n {%- endif %}\n{%- endfor %}\n",
|
| 319 |
+
"clean_up_tokenization_spaces": true,
|
| 320 |
+
"eos_token": "<|endoftext|>",
|
| 321 |
+
"extra_special_tokens": {},
|
| 322 |
+
"model_max_length": 2048,
|
| 323 |
+
"pad_token": "<|endoftext|>",
|
| 324 |
+
"return_token_type_ids": false,
|
| 325 |
+
"tokenizer_class": "CodeGenTokenizer",
|
| 326 |
+
"unk_token": "<|endoftext|>"
|
| 327 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3533 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.051152203381160645,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00010230440676232128,
|
| 13 |
+
"grad_norm": 0.6171875,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"loss": 1.751,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00020460881352464257,
|
| 20 |
+
"grad_norm": 0.6484375,
|
| 21 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 22 |
+
"loss": 1.8722,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0003069132202869639,
|
| 27 |
+
"grad_norm": 0.57421875,
|
| 28 |
+
"learning_rate": 3e-06,
|
| 29 |
+
"loss": 1.5552,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.00040921762704928513,
|
| 34 |
+
"grad_norm": 0.64453125,
|
| 35 |
+
"learning_rate": 4.000000000000001e-06,
|
| 36 |
+
"loss": 1.8481,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0005115220338116064,
|
| 41 |
+
"grad_norm": 0.65625,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 1.6085,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0006138264405739278,
|
| 48 |
+
"grad_norm": 0.6484375,
|
| 49 |
+
"learning_rate": 6e-06,
|
| 50 |
+
"loss": 1.8854,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0007161308473362491,
|
| 55 |
+
"grad_norm": 0.56640625,
|
| 56 |
+
"learning_rate": 7.000000000000001e-06,
|
| 57 |
+
"loss": 1.5885,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0008184352540985703,
|
| 62 |
+
"grad_norm": 0.65234375,
|
| 63 |
+
"learning_rate": 8.000000000000001e-06,
|
| 64 |
+
"loss": 1.6188,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0009207396608608916,
|
| 69 |
+
"grad_norm": 0.58984375,
|
| 70 |
+
"learning_rate": 9e-06,
|
| 71 |
+
"loss": 1.7513,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0010230440676232129,
|
| 76 |
+
"grad_norm": 0.640625,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 1.847,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.001125348474385534,
|
| 83 |
+
"grad_norm": 0.56640625,
|
| 84 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 85 |
+
"loss": 1.5958,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0012276528811478555,
|
| 90 |
+
"grad_norm": 0.55078125,
|
| 91 |
+
"learning_rate": 1.2e-05,
|
| 92 |
+
"loss": 1.7284,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0013299572879101767,
|
| 97 |
+
"grad_norm": 0.58203125,
|
| 98 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 99 |
+
"loss": 1.6829,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0014322616946724981,
|
| 104 |
+
"grad_norm": 0.6484375,
|
| 105 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 106 |
+
"loss": 1.8158,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0015345661014348193,
|
| 111 |
+
"grad_norm": 0.671875,
|
| 112 |
+
"learning_rate": 1.5e-05,
|
| 113 |
+
"loss": 1.9778,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0016368705081971405,
|
| 118 |
+
"grad_norm": 0.609375,
|
| 119 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 120 |
+
"loss": 1.7056,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.001739174914959462,
|
| 125 |
+
"grad_norm": 0.65625,
|
| 126 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 127 |
+
"loss": 1.8864,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0018414793217217831,
|
| 132 |
+
"grad_norm": 0.6484375,
|
| 133 |
+
"learning_rate": 1.8e-05,
|
| 134 |
+
"loss": 1.7853,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0019437837284841043,
|
| 139 |
+
"grad_norm": 0.5859375,
|
| 140 |
+
"learning_rate": 1.9e-05,
|
| 141 |
+
"loss": 1.7343,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0020460881352464258,
|
| 146 |
+
"grad_norm": 0.6796875,
|
| 147 |
+
"learning_rate": 2e-05,
|
| 148 |
+
"loss": 1.9046,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.002148392542008747,
|
| 153 |
+
"grad_norm": 0.609375,
|
| 154 |
+
"learning_rate": 2.1e-05,
|
| 155 |
+
"loss": 1.8652,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.002250696948771068,
|
| 160 |
+
"grad_norm": 0.6171875,
|
| 161 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 162 |
+
"loss": 1.916,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0023530013555333896,
|
| 167 |
+
"grad_norm": 0.63671875,
|
| 168 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 169 |
+
"loss": 1.6101,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.002455305762295711,
|
| 174 |
+
"grad_norm": 0.59375,
|
| 175 |
+
"learning_rate": 2.4e-05,
|
| 176 |
+
"loss": 1.7165,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.002557610169058032,
|
| 181 |
+
"grad_norm": 0.6171875,
|
| 182 |
+
"learning_rate": 2.5e-05,
|
| 183 |
+
"loss": 1.9596,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0026599145758203534,
|
| 188 |
+
"grad_norm": 0.7109375,
|
| 189 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 190 |
+
"loss": 2.0135,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.002762218982582675,
|
| 195 |
+
"grad_norm": 0.59375,
|
| 196 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 197 |
+
"loss": 1.8613,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.0028645233893449962,
|
| 202 |
+
"grad_norm": 0.58203125,
|
| 203 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 204 |
+
"loss": 1.6937,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0029668277961073172,
|
| 209 |
+
"grad_norm": 0.6953125,
|
| 210 |
+
"learning_rate": 2.9e-05,
|
| 211 |
+
"loss": 1.8989,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0030691322028696386,
|
| 216 |
+
"grad_norm": 0.625,
|
| 217 |
+
"learning_rate": 3e-05,
|
| 218 |
+
"loss": 1.9366,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.00317143660963196,
|
| 223 |
+
"grad_norm": 0.62109375,
|
| 224 |
+
"learning_rate": 3.1e-05,
|
| 225 |
+
"loss": 1.8176,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.003273741016394281,
|
| 230 |
+
"grad_norm": 0.7109375,
|
| 231 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 232 |
+
"loss": 1.965,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0033760454231566025,
|
| 237 |
+
"grad_norm": 0.6484375,
|
| 238 |
+
"learning_rate": 3.3e-05,
|
| 239 |
+
"loss": 1.8058,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.003478349829918924,
|
| 244 |
+
"grad_norm": 0.5859375,
|
| 245 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 246 |
+
"loss": 1.8037,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.003580654236681245,
|
| 251 |
+
"grad_norm": 0.6796875,
|
| 252 |
+
"learning_rate": 3.5e-05,
|
| 253 |
+
"loss": 1.9695,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0036829586434435663,
|
| 258 |
+
"grad_norm": 0.609375,
|
| 259 |
+
"learning_rate": 3.6e-05,
|
| 260 |
+
"loss": 1.7983,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0037852630502058877,
|
| 265 |
+
"grad_norm": 0.61328125,
|
| 266 |
+
"learning_rate": 3.7e-05,
|
| 267 |
+
"loss": 1.8376,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.0038875674569682087,
|
| 272 |
+
"grad_norm": 0.56640625,
|
| 273 |
+
"learning_rate": 3.8e-05,
|
| 274 |
+
"loss": 1.6842,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.0039898718637305305,
|
| 279 |
+
"grad_norm": 0.546875,
|
| 280 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 281 |
+
"loss": 1.57,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.0040921762704928515,
|
| 286 |
+
"grad_norm": 0.56640625,
|
| 287 |
+
"learning_rate": 4e-05,
|
| 288 |
+
"loss": 1.8307,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.0041944806772551725,
|
| 293 |
+
"grad_norm": 0.5234375,
|
| 294 |
+
"learning_rate": 4.1e-05,
|
| 295 |
+
"loss": 1.6629,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.004296785084017494,
|
| 300 |
+
"grad_norm": 0.6015625,
|
| 301 |
+
"learning_rate": 4.2e-05,
|
| 302 |
+
"loss": 1.801,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.004399089490779815,
|
| 307 |
+
"grad_norm": 0.546875,
|
| 308 |
+
"learning_rate": 4.3e-05,
|
| 309 |
+
"loss": 1.7171,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.004501393897542136,
|
| 314 |
+
"grad_norm": 0.61328125,
|
| 315 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 316 |
+
"loss": 1.7683,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.004603698304304458,
|
| 321 |
+
"grad_norm": 0.5390625,
|
| 322 |
+
"learning_rate": 4.5e-05,
|
| 323 |
+
"loss": 1.6179,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.004706002711066779,
|
| 328 |
+
"grad_norm": 0.58984375,
|
| 329 |
+
"learning_rate": 4.600000000000001e-05,
|
| 330 |
+
"loss": 1.8107,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.0048083071178291,
|
| 335 |
+
"grad_norm": 0.55859375,
|
| 336 |
+
"learning_rate": 4.7e-05,
|
| 337 |
+
"loss": 1.9287,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.004910611524591422,
|
| 342 |
+
"grad_norm": 0.56640625,
|
| 343 |
+
"learning_rate": 4.8e-05,
|
| 344 |
+
"loss": 1.8884,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.005012915931353743,
|
| 349 |
+
"grad_norm": 0.5625,
|
| 350 |
+
"learning_rate": 4.9e-05,
|
| 351 |
+
"loss": 1.6786,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.005115220338116064,
|
| 356 |
+
"grad_norm": 0.5703125,
|
| 357 |
+
"learning_rate": 5e-05,
|
| 358 |
+
"loss": 1.848,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.005217524744878386,
|
| 363 |
+
"grad_norm": 0.55078125,
|
| 364 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 365 |
+
"loss": 1.6701,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.005319829151640707,
|
| 370 |
+
"grad_norm": 0.57421875,
|
| 371 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 372 |
+
"loss": 1.7677,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.005422133558403028,
|
| 377 |
+
"grad_norm": 0.51171875,
|
| 378 |
+
"learning_rate": 5.300000000000001e-05,
|
| 379 |
+
"loss": 1.5958,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.00552443796516535,
|
| 384 |
+
"grad_norm": 0.59765625,
|
| 385 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 386 |
+
"loss": 1.794,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.005626742371927671,
|
| 391 |
+
"grad_norm": 0.50390625,
|
| 392 |
+
"learning_rate": 5.500000000000001e-05,
|
| 393 |
+
"loss": 1.7429,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.0057290467786899925,
|
| 398 |
+
"grad_norm": 0.61328125,
|
| 399 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 400 |
+
"loss": 1.898,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.0058313511854523135,
|
| 405 |
+
"grad_norm": 0.578125,
|
| 406 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 407 |
+
"loss": 1.9033,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.0059336555922146345,
|
| 412 |
+
"grad_norm": 0.54296875,
|
| 413 |
+
"learning_rate": 5.8e-05,
|
| 414 |
+
"loss": 1.7433,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.006035959998976956,
|
| 419 |
+
"grad_norm": 0.51171875,
|
| 420 |
+
"learning_rate": 5.9e-05,
|
| 421 |
+
"loss": 1.7133,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.006138264405739277,
|
| 426 |
+
"grad_norm": 0.546875,
|
| 427 |
+
"learning_rate": 6e-05,
|
| 428 |
+
"loss": 1.8123,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.006240568812501598,
|
| 433 |
+
"grad_norm": 0.4921875,
|
| 434 |
+
"learning_rate": 6.1e-05,
|
| 435 |
+
"loss": 1.604,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.00634287321926392,
|
| 440 |
+
"grad_norm": 0.55078125,
|
| 441 |
+
"learning_rate": 6.2e-05,
|
| 442 |
+
"loss": 1.7869,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.006445177626026241,
|
| 447 |
+
"grad_norm": 0.439453125,
|
| 448 |
+
"learning_rate": 6.3e-05,
|
| 449 |
+
"loss": 1.559,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.006547482032788562,
|
| 454 |
+
"grad_norm": 0.466796875,
|
| 455 |
+
"learning_rate": 6.400000000000001e-05,
|
| 456 |
+
"loss": 1.7037,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.006649786439550884,
|
| 461 |
+
"grad_norm": 0.6328125,
|
| 462 |
+
"learning_rate": 6.500000000000001e-05,
|
| 463 |
+
"loss": 1.7831,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.006752090846313205,
|
| 468 |
+
"grad_norm": 0.51953125,
|
| 469 |
+
"learning_rate": 6.6e-05,
|
| 470 |
+
"loss": 1.6768,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.006854395253075526,
|
| 475 |
+
"grad_norm": 0.5390625,
|
| 476 |
+
"learning_rate": 6.7e-05,
|
| 477 |
+
"loss": 1.8762,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.006956699659837848,
|
| 482 |
+
"grad_norm": 0.55859375,
|
| 483 |
+
"learning_rate": 6.800000000000001e-05,
|
| 484 |
+
"loss": 1.8037,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.007059004066600169,
|
| 489 |
+
"grad_norm": 0.52734375,
|
| 490 |
+
"learning_rate": 6.9e-05,
|
| 491 |
+
"loss": 1.9943,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.00716130847336249,
|
| 496 |
+
"grad_norm": 0.4453125,
|
| 497 |
+
"learning_rate": 7e-05,
|
| 498 |
+
"loss": 1.6346,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.007263612880124812,
|
| 503 |
+
"grad_norm": 0.52734375,
|
| 504 |
+
"learning_rate": 7.1e-05,
|
| 505 |
+
"loss": 1.7755,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.007365917286887133,
|
| 510 |
+
"grad_norm": 0.494140625,
|
| 511 |
+
"learning_rate": 7.2e-05,
|
| 512 |
+
"loss": 1.9162,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.0074682216936494536,
|
| 517 |
+
"grad_norm": 0.51171875,
|
| 518 |
+
"learning_rate": 7.3e-05,
|
| 519 |
+
"loss": 1.6629,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.007570526100411775,
|
| 524 |
+
"grad_norm": 0.51171875,
|
| 525 |
+
"learning_rate": 7.4e-05,
|
| 526 |
+
"loss": 1.7918,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.007672830507174096,
|
| 531 |
+
"grad_norm": 0.50390625,
|
| 532 |
+
"learning_rate": 7.500000000000001e-05,
|
| 533 |
+
"loss": 1.8483,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.007775134913936417,
|
| 538 |
+
"grad_norm": 0.458984375,
|
| 539 |
+
"learning_rate": 7.6e-05,
|
| 540 |
+
"loss": 1.5213,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.007877439320698738,
|
| 545 |
+
"grad_norm": 0.50390625,
|
| 546 |
+
"learning_rate": 7.7e-05,
|
| 547 |
+
"loss": 1.7234,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.007979743727461061,
|
| 552 |
+
"grad_norm": 0.490234375,
|
| 553 |
+
"learning_rate": 7.800000000000001e-05,
|
| 554 |
+
"loss": 1.5943,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.008082048134223382,
|
| 559 |
+
"grad_norm": 0.51171875,
|
| 560 |
+
"learning_rate": 7.900000000000001e-05,
|
| 561 |
+
"loss": 1.7327,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.008184352540985703,
|
| 566 |
+
"grad_norm": 0.431640625,
|
| 567 |
+
"learning_rate": 8e-05,
|
| 568 |
+
"loss": 1.6263,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.008286656947748024,
|
| 573 |
+
"grad_norm": 0.4765625,
|
| 574 |
+
"learning_rate": 8.1e-05,
|
| 575 |
+
"loss": 1.6922,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.008388961354510345,
|
| 580 |
+
"grad_norm": 0.48828125,
|
| 581 |
+
"learning_rate": 8.2e-05,
|
| 582 |
+
"loss": 1.6937,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.008491265761272666,
|
| 587 |
+
"grad_norm": 0.4375,
|
| 588 |
+
"learning_rate": 8.3e-05,
|
| 589 |
+
"loss": 1.5954,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.008593570168034989,
|
| 594 |
+
"grad_norm": 0.46875,
|
| 595 |
+
"learning_rate": 8.4e-05,
|
| 596 |
+
"loss": 1.7655,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.00869587457479731,
|
| 601 |
+
"grad_norm": 0.48828125,
|
| 602 |
+
"learning_rate": 8.5e-05,
|
| 603 |
+
"loss": 1.8177,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.00879817898155963,
|
| 608 |
+
"grad_norm": 0.466796875,
|
| 609 |
+
"learning_rate": 8.6e-05,
|
| 610 |
+
"loss": 1.6066,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.008900483388321952,
|
| 615 |
+
"grad_norm": 0.4453125,
|
| 616 |
+
"learning_rate": 8.7e-05,
|
| 617 |
+
"loss": 1.7382,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.009002787795084273,
|
| 622 |
+
"grad_norm": 0.4140625,
|
| 623 |
+
"learning_rate": 8.800000000000001e-05,
|
| 624 |
+
"loss": 1.5273,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.009105092201846595,
|
| 629 |
+
"grad_norm": 0.419921875,
|
| 630 |
+
"learning_rate": 8.900000000000001e-05,
|
| 631 |
+
"loss": 1.633,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.009207396608608916,
|
| 636 |
+
"grad_norm": 0.431640625,
|
| 637 |
+
"learning_rate": 9e-05,
|
| 638 |
+
"loss": 1.4586,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.009309701015371237,
|
| 643 |
+
"grad_norm": 0.412109375,
|
| 644 |
+
"learning_rate": 9.1e-05,
|
| 645 |
+
"loss": 1.6444,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.009412005422133558,
|
| 650 |
+
"grad_norm": 0.427734375,
|
| 651 |
+
"learning_rate": 9.200000000000001e-05,
|
| 652 |
+
"loss": 1.6237,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.00951430982889588,
|
| 657 |
+
"grad_norm": 0.498046875,
|
| 658 |
+
"learning_rate": 9.300000000000001e-05,
|
| 659 |
+
"loss": 1.5545,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.0096166142356582,
|
| 664 |
+
"grad_norm": 0.427734375,
|
| 665 |
+
"learning_rate": 9.4e-05,
|
| 666 |
+
"loss": 1.6549,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.009718918642420523,
|
| 671 |
+
"grad_norm": 0.423828125,
|
| 672 |
+
"learning_rate": 9.5e-05,
|
| 673 |
+
"loss": 1.5817,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.009821223049182844,
|
| 678 |
+
"grad_norm": 0.42578125,
|
| 679 |
+
"learning_rate": 9.6e-05,
|
| 680 |
+
"loss": 1.617,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.009923527455945165,
|
| 685 |
+
"grad_norm": 0.474609375,
|
| 686 |
+
"learning_rate": 9.7e-05,
|
| 687 |
+
"loss": 1.7116,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.010025831862707486,
|
| 692 |
+
"grad_norm": 0.40234375,
|
| 693 |
+
"learning_rate": 9.8e-05,
|
| 694 |
+
"loss": 1.6373,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.010128136269469807,
|
| 699 |
+
"grad_norm": 0.37109375,
|
| 700 |
+
"learning_rate": 9.900000000000001e-05,
|
| 701 |
+
"loss": 1.5059,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.010230440676232128,
|
| 706 |
+
"grad_norm": 0.375,
|
| 707 |
+
"learning_rate": 0.0001,
|
| 708 |
+
"loss": 1.4517,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.01033274508299445,
|
| 713 |
+
"grad_norm": 0.3828125,
|
| 714 |
+
"learning_rate": 9.999497487437187e-05,
|
| 715 |
+
"loss": 1.5179,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.010435049489756772,
|
| 720 |
+
"grad_norm": 0.400390625,
|
| 721 |
+
"learning_rate": 9.998994974874373e-05,
|
| 722 |
+
"loss": 1.4234,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.010537353896519093,
|
| 727 |
+
"grad_norm": 0.416015625,
|
| 728 |
+
"learning_rate": 9.998492462311558e-05,
|
| 729 |
+
"loss": 1.7412,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.010639658303281414,
|
| 734 |
+
"grad_norm": 1.6953125,
|
| 735 |
+
"learning_rate": 9.997989949748744e-05,
|
| 736 |
+
"loss": 1.7474,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.010741962710043735,
|
| 741 |
+
"grad_norm": 0.44140625,
|
| 742 |
+
"learning_rate": 9.99748743718593e-05,
|
| 743 |
+
"loss": 1.7978,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.010844267116806056,
|
| 748 |
+
"grad_norm": 0.423828125,
|
| 749 |
+
"learning_rate": 9.996984924623116e-05,
|
| 750 |
+
"loss": 1.5856,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.010946571523568378,
|
| 755 |
+
"grad_norm": 0.455078125,
|
| 756 |
+
"learning_rate": 9.996482412060301e-05,
|
| 757 |
+
"loss": 1.6459,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.0110488759303307,
|
| 762 |
+
"grad_norm": 0.34765625,
|
| 763 |
+
"learning_rate": 9.995979899497487e-05,
|
| 764 |
+
"loss": 2.6578,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.01115118033709302,
|
| 769 |
+
"grad_norm": 0.41796875,
|
| 770 |
+
"learning_rate": 9.995477386934674e-05,
|
| 771 |
+
"loss": 1.5426,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.011253484743855341,
|
| 776 |
+
"grad_norm": 0.35546875,
|
| 777 |
+
"learning_rate": 9.99497487437186e-05,
|
| 778 |
+
"loss": 1.5997,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.011355789150617662,
|
| 783 |
+
"grad_norm": 0.453125,
|
| 784 |
+
"learning_rate": 9.994472361809045e-05,
|
| 785 |
+
"loss": 1.6825,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.011458093557379985,
|
| 790 |
+
"grad_norm": 0.4140625,
|
| 791 |
+
"learning_rate": 9.993969849246232e-05,
|
| 792 |
+
"loss": 1.5822,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.011560397964142306,
|
| 797 |
+
"grad_norm": 0.419921875,
|
| 798 |
+
"learning_rate": 9.993467336683417e-05,
|
| 799 |
+
"loss": 1.6862,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.011662702370904627,
|
| 804 |
+
"grad_norm": 0.41015625,
|
| 805 |
+
"learning_rate": 9.992964824120603e-05,
|
| 806 |
+
"loss": 1.4022,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.011765006777666948,
|
| 811 |
+
"grad_norm": 0.4609375,
|
| 812 |
+
"learning_rate": 9.99246231155779e-05,
|
| 813 |
+
"loss": 1.5491,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.011867311184429269,
|
| 818 |
+
"grad_norm": 0.359375,
|
| 819 |
+
"learning_rate": 9.991959798994976e-05,
|
| 820 |
+
"loss": 1.5846,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.01196961559119159,
|
| 825 |
+
"grad_norm": 0.357421875,
|
| 826 |
+
"learning_rate": 9.99145728643216e-05,
|
| 827 |
+
"loss": 1.5171,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.012071919997953913,
|
| 832 |
+
"grad_norm": 0.37890625,
|
| 833 |
+
"learning_rate": 9.990954773869348e-05,
|
| 834 |
+
"loss": 1.506,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.012174224404716234,
|
| 839 |
+
"grad_norm": 0.392578125,
|
| 840 |
+
"learning_rate": 9.990452261306533e-05,
|
| 841 |
+
"loss": 1.7183,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.012276528811478555,
|
| 846 |
+
"grad_norm": 0.373046875,
|
| 847 |
+
"learning_rate": 9.989949748743719e-05,
|
| 848 |
+
"loss": 1.6749,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.012378833218240876,
|
| 853 |
+
"grad_norm": 0.333984375,
|
| 854 |
+
"learning_rate": 9.989447236180905e-05,
|
| 855 |
+
"loss": 1.6602,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.012481137625003197,
|
| 860 |
+
"grad_norm": 0.37109375,
|
| 861 |
+
"learning_rate": 9.988944723618092e-05,
|
| 862 |
+
"loss": 1.6384,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.012583442031765518,
|
| 867 |
+
"grad_norm": 0.375,
|
| 868 |
+
"learning_rate": 9.988442211055276e-05,
|
| 869 |
+
"loss": 1.6169,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.01268574643852784,
|
| 874 |
+
"grad_norm": 0.380859375,
|
| 875 |
+
"learning_rate": 9.987939698492463e-05,
|
| 876 |
+
"loss": 1.5463,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.012788050845290161,
|
| 881 |
+
"grad_norm": 0.392578125,
|
| 882 |
+
"learning_rate": 9.987437185929649e-05,
|
| 883 |
+
"loss": 1.6794,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.012890355252052482,
|
| 888 |
+
"grad_norm": 0.40625,
|
| 889 |
+
"learning_rate": 9.986934673366835e-05,
|
| 890 |
+
"loss": 1.5976,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.012992659658814803,
|
| 895 |
+
"grad_norm": 0.388671875,
|
| 896 |
+
"learning_rate": 9.98643216080402e-05,
|
| 897 |
+
"loss": 1.5735,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.013094964065577124,
|
| 902 |
+
"grad_norm": 0.435546875,
|
| 903 |
+
"learning_rate": 9.985929648241207e-05,
|
| 904 |
+
"loss": 1.4446,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.013197268472339445,
|
| 909 |
+
"grad_norm": 0.369140625,
|
| 910 |
+
"learning_rate": 9.985427135678392e-05,
|
| 911 |
+
"loss": 1.5231,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.013299572879101768,
|
| 916 |
+
"grad_norm": 0.3515625,
|
| 917 |
+
"learning_rate": 9.984924623115578e-05,
|
| 918 |
+
"loss": 1.6299,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.013401877285864089,
|
| 923 |
+
"grad_norm": 0.384765625,
|
| 924 |
+
"learning_rate": 9.984422110552765e-05,
|
| 925 |
+
"loss": 1.6611,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.01350418169262641,
|
| 930 |
+
"grad_norm": 0.36328125,
|
| 931 |
+
"learning_rate": 9.983919597989951e-05,
|
| 932 |
+
"loss": 1.6093,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.01360648609938873,
|
| 937 |
+
"grad_norm": 0.435546875,
|
| 938 |
+
"learning_rate": 9.983417085427136e-05,
|
| 939 |
+
"loss": 1.5908,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.013708790506151052,
|
| 944 |
+
"grad_norm": 0.3671875,
|
| 945 |
+
"learning_rate": 9.982914572864322e-05,
|
| 946 |
+
"loss": 1.5976,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.013811094912913375,
|
| 951 |
+
"grad_norm": 0.34765625,
|
| 952 |
+
"learning_rate": 9.982412060301508e-05,
|
| 953 |
+
"loss": 1.4695,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.013913399319675696,
|
| 958 |
+
"grad_norm": 0.345703125,
|
| 959 |
+
"learning_rate": 9.981909547738694e-05,
|
| 960 |
+
"loss": 1.4971,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.014015703726438017,
|
| 965 |
+
"grad_norm": 0.37109375,
|
| 966 |
+
"learning_rate": 9.98140703517588e-05,
|
| 967 |
+
"loss": 1.6059,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.014118008133200338,
|
| 972 |
+
"grad_norm": 0.333984375,
|
| 973 |
+
"learning_rate": 9.980904522613065e-05,
|
| 974 |
+
"loss": 1.5415,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.014220312539962658,
|
| 979 |
+
"grad_norm": 0.404296875,
|
| 980 |
+
"learning_rate": 9.980402010050252e-05,
|
| 981 |
+
"loss": 1.6073,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.01432261694672498,
|
| 986 |
+
"grad_norm": 0.359375,
|
| 987 |
+
"learning_rate": 9.979899497487438e-05,
|
| 988 |
+
"loss": 1.4738,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.014424921353487302,
|
| 993 |
+
"grad_norm": 0.3515625,
|
| 994 |
+
"learning_rate": 9.979396984924624e-05,
|
| 995 |
+
"loss": 1.6089,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.014527225760249623,
|
| 1000 |
+
"grad_norm": 0.34765625,
|
| 1001 |
+
"learning_rate": 9.978894472361809e-05,
|
| 1002 |
+
"loss": 1.5486,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.014629530167011944,
|
| 1007 |
+
"grad_norm": 0.31640625,
|
| 1008 |
+
"learning_rate": 9.978391959798995e-05,
|
| 1009 |
+
"loss": 1.5261,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.014731834573774265,
|
| 1014 |
+
"grad_norm": 0.3984375,
|
| 1015 |
+
"learning_rate": 9.977889447236181e-05,
|
| 1016 |
+
"loss": 1.726,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.014834138980536586,
|
| 1021 |
+
"grad_norm": 0.380859375,
|
| 1022 |
+
"learning_rate": 9.977386934673367e-05,
|
| 1023 |
+
"loss": 1.7616,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.014936443387298907,
|
| 1028 |
+
"grad_norm": 0.3671875,
|
| 1029 |
+
"learning_rate": 9.976884422110552e-05,
|
| 1030 |
+
"loss": 1.8157,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.01503874779406123,
|
| 1035 |
+
"grad_norm": 0.37109375,
|
| 1036 |
+
"learning_rate": 9.97638190954774e-05,
|
| 1037 |
+
"loss": 1.5536,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.01514105220082355,
|
| 1042 |
+
"grad_norm": 0.373046875,
|
| 1043 |
+
"learning_rate": 9.975879396984925e-05,
|
| 1044 |
+
"loss": 1.6118,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.015243356607585872,
|
| 1049 |
+
"grad_norm": 0.357421875,
|
| 1050 |
+
"learning_rate": 9.975376884422111e-05,
|
| 1051 |
+
"loss": 1.7143,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.015345661014348193,
|
| 1056 |
+
"grad_norm": 0.345703125,
|
| 1057 |
+
"learning_rate": 9.974874371859297e-05,
|
| 1058 |
+
"loss": 1.4682,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.015447965421110514,
|
| 1063 |
+
"grad_norm": 0.33203125,
|
| 1064 |
+
"learning_rate": 9.974371859296483e-05,
|
| 1065 |
+
"loss": 1.5663,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.015550269827872835,
|
| 1070 |
+
"grad_norm": 0.353515625,
|
| 1071 |
+
"learning_rate": 9.973869346733668e-05,
|
| 1072 |
+
"loss": 1.6771,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.015652574234635157,
|
| 1077 |
+
"grad_norm": 0.359375,
|
| 1078 |
+
"learning_rate": 9.973366834170856e-05,
|
| 1079 |
+
"loss": 1.6855,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.015754878641397477,
|
| 1084 |
+
"grad_norm": 0.47265625,
|
| 1085 |
+
"learning_rate": 9.97286432160804e-05,
|
| 1086 |
+
"loss": 1.7027,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.0158571830481598,
|
| 1091 |
+
"grad_norm": 0.3359375,
|
| 1092 |
+
"learning_rate": 9.972361809045227e-05,
|
| 1093 |
+
"loss": 1.4748,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.015959487454922122,
|
| 1098 |
+
"grad_norm": 0.349609375,
|
| 1099 |
+
"learning_rate": 9.971859296482412e-05,
|
| 1100 |
+
"loss": 1.5864,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.01606179186168444,
|
| 1105 |
+
"grad_norm": 0.33984375,
|
| 1106 |
+
"learning_rate": 9.971356783919599e-05,
|
| 1107 |
+
"loss": 1.5928,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.016164096268446764,
|
| 1112 |
+
"grad_norm": 0.341796875,
|
| 1113 |
+
"learning_rate": 9.970854271356784e-05,
|
| 1114 |
+
"loss": 1.7351,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.016266400675209083,
|
| 1119 |
+
"grad_norm": 0.37890625,
|
| 1120 |
+
"learning_rate": 9.97035175879397e-05,
|
| 1121 |
+
"loss": 1.7008,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.016368705081971406,
|
| 1126 |
+
"grad_norm": 0.34375,
|
| 1127 |
+
"learning_rate": 9.969849246231156e-05,
|
| 1128 |
+
"loss": 1.4992,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.01647100948873373,
|
| 1133 |
+
"grad_norm": 0.380859375,
|
| 1134 |
+
"learning_rate": 9.969346733668343e-05,
|
| 1135 |
+
"loss": 1.635,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.016573313895496048,
|
| 1140 |
+
"grad_norm": 0.353515625,
|
| 1141 |
+
"learning_rate": 9.968844221105527e-05,
|
| 1142 |
+
"loss": 1.5781,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.01667561830225837,
|
| 1147 |
+
"grad_norm": 0.3515625,
|
| 1148 |
+
"learning_rate": 9.968341708542715e-05,
|
| 1149 |
+
"loss": 1.4841,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.01677792270902069,
|
| 1154 |
+
"grad_norm": 0.3203125,
|
| 1155 |
+
"learning_rate": 9.9678391959799e-05,
|
| 1156 |
+
"loss": 1.3593,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.016880227115783013,
|
| 1161 |
+
"grad_norm": 0.359375,
|
| 1162 |
+
"learning_rate": 9.967336683417086e-05,
|
| 1163 |
+
"loss": 1.5774,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.016982531522545332,
|
| 1168 |
+
"grad_norm": 0.326171875,
|
| 1169 |
+
"learning_rate": 9.966834170854272e-05,
|
| 1170 |
+
"loss": 1.5977,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.017084835929307655,
|
| 1175 |
+
"grad_norm": 0.3359375,
|
| 1176 |
+
"learning_rate": 9.966331658291458e-05,
|
| 1177 |
+
"loss": 1.4401,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.017187140336069977,
|
| 1182 |
+
"grad_norm": 0.45703125,
|
| 1183 |
+
"learning_rate": 9.965829145728643e-05,
|
| 1184 |
+
"loss": 1.4486,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.017289444742832297,
|
| 1189 |
+
"grad_norm": 0.3828125,
|
| 1190 |
+
"learning_rate": 9.96532663316583e-05,
|
| 1191 |
+
"loss": 1.5923,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.01739174914959462,
|
| 1196 |
+
"grad_norm": 0.330078125,
|
| 1197 |
+
"learning_rate": 9.964824120603016e-05,
|
| 1198 |
+
"loss": 1.6353,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.01749405355635694,
|
| 1203 |
+
"grad_norm": 0.33203125,
|
| 1204 |
+
"learning_rate": 9.964321608040202e-05,
|
| 1205 |
+
"loss": 1.4871,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.01759635796311926,
|
| 1210 |
+
"grad_norm": 0.36328125,
|
| 1211 |
+
"learning_rate": 9.963819095477387e-05,
|
| 1212 |
+
"loss": 1.6157,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.017698662369881584,
|
| 1217 |
+
"grad_norm": 0.32421875,
|
| 1218 |
+
"learning_rate": 9.963316582914573e-05,
|
| 1219 |
+
"loss": 1.5152,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.017800966776643903,
|
| 1224 |
+
"grad_norm": 0.333984375,
|
| 1225 |
+
"learning_rate": 9.962814070351759e-05,
|
| 1226 |
+
"loss": 1.5647,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.017903271183406226,
|
| 1231 |
+
"grad_norm": 0.359375,
|
| 1232 |
+
"learning_rate": 9.962311557788945e-05,
|
| 1233 |
+
"loss": 1.6144,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.018005575590168545,
|
| 1238 |
+
"grad_norm": 0.326171875,
|
| 1239 |
+
"learning_rate": 9.961809045226132e-05,
|
| 1240 |
+
"loss": 1.5505,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.018107879996930868,
|
| 1245 |
+
"grad_norm": 0.318359375,
|
| 1246 |
+
"learning_rate": 9.961306532663316e-05,
|
| 1247 |
+
"loss": 1.4726,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.01821018440369319,
|
| 1252 |
+
"grad_norm": 0.330078125,
|
| 1253 |
+
"learning_rate": 9.960804020100503e-05,
|
| 1254 |
+
"loss": 1.484,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.01831248881045551,
|
| 1259 |
+
"grad_norm": 0.34765625,
|
| 1260 |
+
"learning_rate": 9.960301507537689e-05,
|
| 1261 |
+
"loss": 1.3879,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.018414793217217833,
|
| 1266 |
+
"grad_norm": 0.369140625,
|
| 1267 |
+
"learning_rate": 9.959798994974875e-05,
|
| 1268 |
+
"loss": 1.5664,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.018517097623980152,
|
| 1273 |
+
"grad_norm": 0.3359375,
|
| 1274 |
+
"learning_rate": 9.95929648241206e-05,
|
| 1275 |
+
"loss": 1.5777,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.018619402030742475,
|
| 1280 |
+
"grad_norm": 0.353515625,
|
| 1281 |
+
"learning_rate": 9.958793969849247e-05,
|
| 1282 |
+
"loss": 1.4148,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.018721706437504794,
|
| 1287 |
+
"grad_norm": 0.3671875,
|
| 1288 |
+
"learning_rate": 9.958291457286432e-05,
|
| 1289 |
+
"loss": 1.5065,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.018824010844267117,
|
| 1294 |
+
"grad_norm": 0.328125,
|
| 1295 |
+
"learning_rate": 9.957788944723619e-05,
|
| 1296 |
+
"loss": 1.5637,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.01892631525102944,
|
| 1301 |
+
"grad_norm": 0.314453125,
|
| 1302 |
+
"learning_rate": 9.957286432160805e-05,
|
| 1303 |
+
"loss": 1.4061,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.01902861965779176,
|
| 1308 |
+
"grad_norm": 0.333984375,
|
| 1309 |
+
"learning_rate": 9.956783919597991e-05,
|
| 1310 |
+
"loss": 1.5302,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.01913092406455408,
|
| 1315 |
+
"grad_norm": 0.330078125,
|
| 1316 |
+
"learning_rate": 9.956281407035176e-05,
|
| 1317 |
+
"loss": 1.5825,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.0192332284713164,
|
| 1322 |
+
"grad_norm": 0.2890625,
|
| 1323 |
+
"learning_rate": 9.955778894472362e-05,
|
| 1324 |
+
"loss": 1.4345,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.019335532878078723,
|
| 1329 |
+
"grad_norm": 0.32421875,
|
| 1330 |
+
"learning_rate": 9.955276381909548e-05,
|
| 1331 |
+
"loss": 1.4496,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.019437837284841046,
|
| 1336 |
+
"grad_norm": 0.326171875,
|
| 1337 |
+
"learning_rate": 9.954773869346734e-05,
|
| 1338 |
+
"loss": 1.5227,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.019540141691603365,
|
| 1343 |
+
"grad_norm": 0.3203125,
|
| 1344 |
+
"learning_rate": 9.954271356783919e-05,
|
| 1345 |
+
"loss": 1.5828,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.019642446098365688,
|
| 1350 |
+
"grad_norm": 0.32421875,
|
| 1351 |
+
"learning_rate": 9.953768844221107e-05,
|
| 1352 |
+
"loss": 1.4511,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.019744750505128007,
|
| 1357 |
+
"grad_norm": 0.34375,
|
| 1358 |
+
"learning_rate": 9.953266331658292e-05,
|
| 1359 |
+
"loss": 1.4965,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.01984705491189033,
|
| 1364 |
+
"grad_norm": 0.341796875,
|
| 1365 |
+
"learning_rate": 9.952763819095478e-05,
|
| 1366 |
+
"loss": 1.6729,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.01994935931865265,
|
| 1371 |
+
"grad_norm": 0.33984375,
|
| 1372 |
+
"learning_rate": 9.952261306532664e-05,
|
| 1373 |
+
"loss": 1.7777,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.020051663725414972,
|
| 1378 |
+
"grad_norm": 0.330078125,
|
| 1379 |
+
"learning_rate": 9.95175879396985e-05,
|
| 1380 |
+
"loss": 1.4149,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.020153968132177295,
|
| 1385 |
+
"grad_norm": 0.337890625,
|
| 1386 |
+
"learning_rate": 9.951256281407035e-05,
|
| 1387 |
+
"loss": 1.7133,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.020256272538939614,
|
| 1392 |
+
"grad_norm": 0.3515625,
|
| 1393 |
+
"learning_rate": 9.950753768844223e-05,
|
| 1394 |
+
"loss": 1.5531,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.020358576945701937,
|
| 1399 |
+
"grad_norm": 0.3671875,
|
| 1400 |
+
"learning_rate": 9.950251256281408e-05,
|
| 1401 |
+
"loss": 1.6573,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.020460881352464256,
|
| 1406 |
+
"grad_norm": 0.33203125,
|
| 1407 |
+
"learning_rate": 9.949748743718594e-05,
|
| 1408 |
+
"loss": 1.3231,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.02056318575922658,
|
| 1413 |
+
"grad_norm": 0.337890625,
|
| 1414 |
+
"learning_rate": 9.94924623115578e-05,
|
| 1415 |
+
"loss": 1.7199,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.0206654901659889,
|
| 1420 |
+
"grad_norm": 0.326171875,
|
| 1421 |
+
"learning_rate": 9.948743718592966e-05,
|
| 1422 |
+
"loss": 1.4348,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.02076779457275122,
|
| 1427 |
+
"grad_norm": 0.3203125,
|
| 1428 |
+
"learning_rate": 9.948241206030151e-05,
|
| 1429 |
+
"loss": 1.51,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.020870098979513543,
|
| 1434 |
+
"grad_norm": 0.3125,
|
| 1435 |
+
"learning_rate": 9.947738693467337e-05,
|
| 1436 |
+
"loss": 1.4171,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.020972403386275863,
|
| 1441 |
+
"grad_norm": 0.326171875,
|
| 1442 |
+
"learning_rate": 9.947236180904523e-05,
|
| 1443 |
+
"loss": 1.5366,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.021074707793038185,
|
| 1448 |
+
"grad_norm": 0.310546875,
|
| 1449 |
+
"learning_rate": 9.94673366834171e-05,
|
| 1450 |
+
"loss": 1.5415,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.021177012199800508,
|
| 1455 |
+
"grad_norm": 0.31640625,
|
| 1456 |
+
"learning_rate": 9.946231155778894e-05,
|
| 1457 |
+
"loss": 1.5641,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.021279316606562827,
|
| 1462 |
+
"grad_norm": 0.31640625,
|
| 1463 |
+
"learning_rate": 9.94572864321608e-05,
|
| 1464 |
+
"loss": 1.5054,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.02138162101332515,
|
| 1469 |
+
"grad_norm": 0.34375,
|
| 1470 |
+
"learning_rate": 9.945226130653267e-05,
|
| 1471 |
+
"loss": 1.5638,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.02148392542008747,
|
| 1476 |
+
"grad_norm": 0.318359375,
|
| 1477 |
+
"learning_rate": 9.944723618090453e-05,
|
| 1478 |
+
"loss": 1.4641,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.021586229826849792,
|
| 1483 |
+
"grad_norm": 0.306640625,
|
| 1484 |
+
"learning_rate": 9.944221105527639e-05,
|
| 1485 |
+
"loss": 1.5075,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.02168853423361211,
|
| 1490 |
+
"grad_norm": 0.318359375,
|
| 1491 |
+
"learning_rate": 9.943718592964824e-05,
|
| 1492 |
+
"loss": 1.3774,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.021790838640374434,
|
| 1497 |
+
"grad_norm": 0.326171875,
|
| 1498 |
+
"learning_rate": 9.94321608040201e-05,
|
| 1499 |
+
"loss": 1.5137,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.021893143047136757,
|
| 1504 |
+
"grad_norm": 0.3203125,
|
| 1505 |
+
"learning_rate": 9.942713567839197e-05,
|
| 1506 |
+
"loss": 1.463,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.021995447453899076,
|
| 1511 |
+
"grad_norm": 0.3359375,
|
| 1512 |
+
"learning_rate": 9.942211055276383e-05,
|
| 1513 |
+
"loss": 1.5722,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.0220977518606614,
|
| 1518 |
+
"grad_norm": 0.3203125,
|
| 1519 |
+
"learning_rate": 9.941708542713568e-05,
|
| 1520 |
+
"loss": 1.5695,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.022200056267423718,
|
| 1525 |
+
"grad_norm": 0.306640625,
|
| 1526 |
+
"learning_rate": 9.941206030150754e-05,
|
| 1527 |
+
"loss": 1.4522,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.02230236067418604,
|
| 1532 |
+
"grad_norm": 0.314453125,
|
| 1533 |
+
"learning_rate": 9.94070351758794e-05,
|
| 1534 |
+
"loss": 1.5649,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.022404665080948363,
|
| 1539 |
+
"grad_norm": 0.30078125,
|
| 1540 |
+
"learning_rate": 9.940201005025126e-05,
|
| 1541 |
+
"loss": 1.3819,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.022506969487710683,
|
| 1546 |
+
"grad_norm": 0.31640625,
|
| 1547 |
+
"learning_rate": 9.939698492462311e-05,
|
| 1548 |
+
"loss": 1.4209,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.022609273894473005,
|
| 1553 |
+
"grad_norm": 0.3125,
|
| 1554 |
+
"learning_rate": 9.939195979899499e-05,
|
| 1555 |
+
"loss": 1.5313,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.022711578301235325,
|
| 1560 |
+
"grad_norm": 0.294921875,
|
| 1561 |
+
"learning_rate": 9.938693467336683e-05,
|
| 1562 |
+
"loss": 1.6329,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.022813882707997647,
|
| 1567 |
+
"grad_norm": 0.294921875,
|
| 1568 |
+
"learning_rate": 9.93819095477387e-05,
|
| 1569 |
+
"loss": 1.5602,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.02291618711475997,
|
| 1574 |
+
"grad_norm": 0.30859375,
|
| 1575 |
+
"learning_rate": 9.937688442211056e-05,
|
| 1576 |
+
"loss": 1.4092,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.02301849152152229,
|
| 1581 |
+
"grad_norm": 0.306640625,
|
| 1582 |
+
"learning_rate": 9.937185929648242e-05,
|
| 1583 |
+
"loss": 1.507,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.023120795928284612,
|
| 1588 |
+
"grad_norm": 0.3125,
|
| 1589 |
+
"learning_rate": 9.936683417085427e-05,
|
| 1590 |
+
"loss": 1.3996,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.02322310033504693,
|
| 1595 |
+
"grad_norm": 0.294921875,
|
| 1596 |
+
"learning_rate": 9.936180904522614e-05,
|
| 1597 |
+
"loss": 1.3146,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.023325404741809254,
|
| 1602 |
+
"grad_norm": 0.30859375,
|
| 1603 |
+
"learning_rate": 9.935678391959799e-05,
|
| 1604 |
+
"loss": 1.4642,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.023427709148571573,
|
| 1609 |
+
"grad_norm": 0.29296875,
|
| 1610 |
+
"learning_rate": 9.935175879396985e-05,
|
| 1611 |
+
"loss": 1.5534,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.023530013555333896,
|
| 1616 |
+
"grad_norm": 0.294921875,
|
| 1617 |
+
"learning_rate": 9.934673366834172e-05,
|
| 1618 |
+
"loss": 1.5299,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.02363231796209622,
|
| 1623 |
+
"grad_norm": 0.31640625,
|
| 1624 |
+
"learning_rate": 9.934170854271358e-05,
|
| 1625 |
+
"loss": 1.6027,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.023734622368858538,
|
| 1630 |
+
"grad_norm": 0.3046875,
|
| 1631 |
+
"learning_rate": 9.933668341708543e-05,
|
| 1632 |
+
"loss": 1.3886,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.02383692677562086,
|
| 1637 |
+
"grad_norm": 0.337890625,
|
| 1638 |
+
"learning_rate": 9.933165829145729e-05,
|
| 1639 |
+
"loss": 1.513,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.02393923118238318,
|
| 1644 |
+
"grad_norm": 0.3203125,
|
| 1645 |
+
"learning_rate": 9.932663316582915e-05,
|
| 1646 |
+
"loss": 1.5952,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.024041535589145503,
|
| 1651 |
+
"grad_norm": 0.330078125,
|
| 1652 |
+
"learning_rate": 9.932160804020101e-05,
|
| 1653 |
+
"loss": 1.4902,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.024143839995907825,
|
| 1658 |
+
"grad_norm": 0.314453125,
|
| 1659 |
+
"learning_rate": 9.931658291457286e-05,
|
| 1660 |
+
"loss": 1.5195,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.024246144402670144,
|
| 1665 |
+
"grad_norm": 0.376953125,
|
| 1666 |
+
"learning_rate": 9.931155778894474e-05,
|
| 1667 |
+
"loss": 1.5003,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.024348448809432467,
|
| 1672 |
+
"grad_norm": 0.306640625,
|
| 1673 |
+
"learning_rate": 9.930653266331659e-05,
|
| 1674 |
+
"loss": 1.6125,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.024450753216194786,
|
| 1679 |
+
"grad_norm": 0.3359375,
|
| 1680 |
+
"learning_rate": 9.930150753768845e-05,
|
| 1681 |
+
"loss": 1.4048,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.02455305762295711,
|
| 1686 |
+
"grad_norm": 0.3359375,
|
| 1687 |
+
"learning_rate": 9.929648241206031e-05,
|
| 1688 |
+
"loss": 1.404,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.024655362029719432,
|
| 1693 |
+
"grad_norm": 0.30859375,
|
| 1694 |
+
"learning_rate": 9.929145728643217e-05,
|
| 1695 |
+
"loss": 1.5336,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.02475766643648175,
|
| 1700 |
+
"grad_norm": 0.296875,
|
| 1701 |
+
"learning_rate": 9.928643216080402e-05,
|
| 1702 |
+
"loss": 1.5893,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.024859970843244074,
|
| 1707 |
+
"grad_norm": 0.29296875,
|
| 1708 |
+
"learning_rate": 9.928140703517588e-05,
|
| 1709 |
+
"loss": 1.4552,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.024962275250006393,
|
| 1714 |
+
"grad_norm": 0.31640625,
|
| 1715 |
+
"learning_rate": 9.927638190954774e-05,
|
| 1716 |
+
"loss": 1.5381,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.025064579656768716,
|
| 1721 |
+
"grad_norm": 0.283203125,
|
| 1722 |
+
"learning_rate": 9.927135678391961e-05,
|
| 1723 |
+
"loss": 1.4568,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.025166884063531035,
|
| 1728 |
+
"grad_norm": 0.275390625,
|
| 1729 |
+
"learning_rate": 9.926633165829147e-05,
|
| 1730 |
+
"loss": 1.3433,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.025269188470293358,
|
| 1735 |
+
"grad_norm": 0.30078125,
|
| 1736 |
+
"learning_rate": 9.926130653266332e-05,
|
| 1737 |
+
"loss": 1.4223,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.02537149287705568,
|
| 1742 |
+
"grad_norm": 0.310546875,
|
| 1743 |
+
"learning_rate": 9.925628140703518e-05,
|
| 1744 |
+
"loss": 1.6224,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.025473797283818,
|
| 1749 |
+
"grad_norm": 0.298828125,
|
| 1750 |
+
"learning_rate": 9.925125628140703e-05,
|
| 1751 |
+
"loss": 1.4886,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.025576101690580322,
|
| 1756 |
+
"grad_norm": 0.32421875,
|
| 1757 |
+
"learning_rate": 9.92462311557789e-05,
|
| 1758 |
+
"loss": 1.3823,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.02567840609734264,
|
| 1763 |
+
"grad_norm": 0.322265625,
|
| 1764 |
+
"learning_rate": 9.924120603015075e-05,
|
| 1765 |
+
"loss": 1.6174,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.025780710504104964,
|
| 1770 |
+
"grad_norm": 0.31640625,
|
| 1771 |
+
"learning_rate": 9.923618090452261e-05,
|
| 1772 |
+
"loss": 1.6667,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.025883014910867287,
|
| 1777 |
+
"grad_norm": 0.30859375,
|
| 1778 |
+
"learning_rate": 9.923115577889448e-05,
|
| 1779 |
+
"loss": 1.5771,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.025985319317629606,
|
| 1784 |
+
"grad_norm": 0.2890625,
|
| 1785 |
+
"learning_rate": 9.922613065326634e-05,
|
| 1786 |
+
"loss": 1.3376,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.02608762372439193,
|
| 1791 |
+
"grad_norm": 0.349609375,
|
| 1792 |
+
"learning_rate": 9.922110552763819e-05,
|
| 1793 |
+
"loss": 1.621,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.02618992813115425,
|
| 1798 |
+
"grad_norm": 0.27734375,
|
| 1799 |
+
"learning_rate": 9.921608040201006e-05,
|
| 1800 |
+
"loss": 1.413,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.02629223253791657,
|
| 1805 |
+
"grad_norm": 0.294921875,
|
| 1806 |
+
"learning_rate": 9.921105527638191e-05,
|
| 1807 |
+
"loss": 1.6237,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.02639453694467889,
|
| 1812 |
+
"grad_norm": 0.349609375,
|
| 1813 |
+
"learning_rate": 9.920603015075377e-05,
|
| 1814 |
+
"loss": 1.459,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.026496841351441213,
|
| 1819 |
+
"grad_norm": 0.306640625,
|
| 1820 |
+
"learning_rate": 9.920100502512563e-05,
|
| 1821 |
+
"loss": 1.5026,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.026599145758203536,
|
| 1826 |
+
"grad_norm": 0.314453125,
|
| 1827 |
+
"learning_rate": 9.91959798994975e-05,
|
| 1828 |
+
"loss": 1.517,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.026701450164965855,
|
| 1833 |
+
"grad_norm": 0.373046875,
|
| 1834 |
+
"learning_rate": 9.919095477386935e-05,
|
| 1835 |
+
"loss": 1.6779,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.026803754571728178,
|
| 1840 |
+
"grad_norm": 0.322265625,
|
| 1841 |
+
"learning_rate": 9.918592964824122e-05,
|
| 1842 |
+
"loss": 1.4558,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.026906058978490497,
|
| 1847 |
+
"grad_norm": 0.287109375,
|
| 1848 |
+
"learning_rate": 9.918090452261307e-05,
|
| 1849 |
+
"loss": 1.451,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.02700836338525282,
|
| 1854 |
+
"grad_norm": 0.28515625,
|
| 1855 |
+
"learning_rate": 9.917587939698493e-05,
|
| 1856 |
+
"loss": 1.538,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.027110667792015142,
|
| 1861 |
+
"grad_norm": 0.294921875,
|
| 1862 |
+
"learning_rate": 9.917085427135678e-05,
|
| 1863 |
+
"loss": 1.4029,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.02721297219877746,
|
| 1868 |
+
"grad_norm": 0.302734375,
|
| 1869 |
+
"learning_rate": 9.916582914572866e-05,
|
| 1870 |
+
"loss": 1.6634,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.027315276605539784,
|
| 1875 |
+
"grad_norm": 0.31640625,
|
| 1876 |
+
"learning_rate": 9.91608040201005e-05,
|
| 1877 |
+
"loss": 1.4518,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.027417581012302104,
|
| 1882 |
+
"grad_norm": 0.322265625,
|
| 1883 |
+
"learning_rate": 9.915577889447237e-05,
|
| 1884 |
+
"loss": 1.5737,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.027519885419064426,
|
| 1889 |
+
"grad_norm": 0.310546875,
|
| 1890 |
+
"learning_rate": 9.915075376884423e-05,
|
| 1891 |
+
"loss": 1.4141,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.02762218982582675,
|
| 1896 |
+
"grad_norm": 0.3046875,
|
| 1897 |
+
"learning_rate": 9.914572864321609e-05,
|
| 1898 |
+
"loss": 1.6944,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.02772449423258907,
|
| 1903 |
+
"grad_norm": 0.40234375,
|
| 1904 |
+
"learning_rate": 9.914070351758794e-05,
|
| 1905 |
+
"loss": 1.6906,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.02782679863935139,
|
| 1910 |
+
"grad_norm": 0.322265625,
|
| 1911 |
+
"learning_rate": 9.913567839195981e-05,
|
| 1912 |
+
"loss": 1.7094,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.02792910304611371,
|
| 1917 |
+
"grad_norm": 0.3125,
|
| 1918 |
+
"learning_rate": 9.913065326633166e-05,
|
| 1919 |
+
"loss": 1.5747,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.028031407452876033,
|
| 1924 |
+
"grad_norm": 0.326171875,
|
| 1925 |
+
"learning_rate": 9.912562814070352e-05,
|
| 1926 |
+
"loss": 1.508,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.028133711859638352,
|
| 1931 |
+
"grad_norm": 0.287109375,
|
| 1932 |
+
"learning_rate": 9.912060301507539e-05,
|
| 1933 |
+
"loss": 1.4204,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.028236016266400675,
|
| 1938 |
+
"grad_norm": 0.3046875,
|
| 1939 |
+
"learning_rate": 9.911557788944725e-05,
|
| 1940 |
+
"loss": 1.3767,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.028338320673162998,
|
| 1945 |
+
"grad_norm": 0.296875,
|
| 1946 |
+
"learning_rate": 9.91105527638191e-05,
|
| 1947 |
+
"loss": 1.5665,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.028440625079925317,
|
| 1952 |
+
"grad_norm": 0.287109375,
|
| 1953 |
+
"learning_rate": 9.910552763819096e-05,
|
| 1954 |
+
"loss": 1.4584,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.02854292948668764,
|
| 1959 |
+
"grad_norm": 0.275390625,
|
| 1960 |
+
"learning_rate": 9.910050251256282e-05,
|
| 1961 |
+
"loss": 1.5575,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.02864523389344996,
|
| 1966 |
+
"grad_norm": 0.29296875,
|
| 1967 |
+
"learning_rate": 9.909547738693468e-05,
|
| 1968 |
+
"loss": 1.6704,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.02874753830021228,
|
| 1973 |
+
"grad_norm": 0.310546875,
|
| 1974 |
+
"learning_rate": 9.909045226130653e-05,
|
| 1975 |
+
"loss": 1.6166,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.028849842706974604,
|
| 1980 |
+
"grad_norm": 0.341796875,
|
| 1981 |
+
"learning_rate": 9.90854271356784e-05,
|
| 1982 |
+
"loss": 1.5372,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.028952147113736924,
|
| 1987 |
+
"grad_norm": 0.314453125,
|
| 1988 |
+
"learning_rate": 9.908040201005026e-05,
|
| 1989 |
+
"loss": 1.5092,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.029054451520499246,
|
| 1994 |
+
"grad_norm": 0.302734375,
|
| 1995 |
+
"learning_rate": 9.90753768844221e-05,
|
| 1996 |
+
"loss": 1.5686,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.029156755927261566,
|
| 2001 |
+
"grad_norm": 0.30859375,
|
| 2002 |
+
"learning_rate": 9.907035175879398e-05,
|
| 2003 |
+
"loss": 1.4031,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.02925906033402389,
|
| 2008 |
+
"grad_norm": 0.294921875,
|
| 2009 |
+
"learning_rate": 9.906532663316583e-05,
|
| 2010 |
+
"loss": 1.4016,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.02936136474078621,
|
| 2015 |
+
"grad_norm": 0.294921875,
|
| 2016 |
+
"learning_rate": 9.906030150753769e-05,
|
| 2017 |
+
"loss": 1.4951,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.02946366914754853,
|
| 2022 |
+
"grad_norm": 0.3203125,
|
| 2023 |
+
"learning_rate": 9.905527638190955e-05,
|
| 2024 |
+
"loss": 1.4838,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.029565973554310853,
|
| 2029 |
+
"grad_norm": 0.3359375,
|
| 2030 |
+
"learning_rate": 9.905025125628141e-05,
|
| 2031 |
+
"loss": 1.6449,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.029668277961073172,
|
| 2036 |
+
"grad_norm": 0.302734375,
|
| 2037 |
+
"learning_rate": 9.904522613065326e-05,
|
| 2038 |
+
"loss": 1.6668,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.029770582367835495,
|
| 2043 |
+
"grad_norm": 0.283203125,
|
| 2044 |
+
"learning_rate": 9.904020100502514e-05,
|
| 2045 |
+
"loss": 1.4133,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.029872886774597814,
|
| 2050 |
+
"grad_norm": 0.296875,
|
| 2051 |
+
"learning_rate": 9.903517587939699e-05,
|
| 2052 |
+
"loss": 1.5274,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.029975191181360137,
|
| 2057 |
+
"grad_norm": 0.28515625,
|
| 2058 |
+
"learning_rate": 9.903015075376885e-05,
|
| 2059 |
+
"loss": 1.4883,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.03007749558812246,
|
| 2064 |
+
"grad_norm": 0.32421875,
|
| 2065 |
+
"learning_rate": 9.902512562814071e-05,
|
| 2066 |
+
"loss": 1.5292,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.03017979999488478,
|
| 2071 |
+
"grad_norm": 0.390625,
|
| 2072 |
+
"learning_rate": 9.902010050251257e-05,
|
| 2073 |
+
"loss": 1.5329,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.0302821044016471,
|
| 2078 |
+
"grad_norm": 0.29296875,
|
| 2079 |
+
"learning_rate": 9.901507537688442e-05,
|
| 2080 |
+
"loss": 1.5716,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.03038440880840942,
|
| 2085 |
+
"grad_norm": 0.310546875,
|
| 2086 |
+
"learning_rate": 9.901005025125628e-05,
|
| 2087 |
+
"loss": 1.6152,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.030486713215171744,
|
| 2092 |
+
"grad_norm": 0.287109375,
|
| 2093 |
+
"learning_rate": 9.900502512562815e-05,
|
| 2094 |
+
"loss": 1.5215,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.030589017621934066,
|
| 2099 |
+
"grad_norm": 0.26171875,
|
| 2100 |
+
"learning_rate": 9.900000000000001e-05,
|
| 2101 |
+
"loss": 1.4974,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.030691322028696386,
|
| 2106 |
+
"grad_norm": 0.294921875,
|
| 2107 |
+
"learning_rate": 9.899497487437186e-05,
|
| 2108 |
+
"loss": 1.429,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.03079362643545871,
|
| 2113 |
+
"grad_norm": 0.31640625,
|
| 2114 |
+
"learning_rate": 9.898994974874373e-05,
|
| 2115 |
+
"loss": 1.6325,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.030895930842221028,
|
| 2120 |
+
"grad_norm": 0.26171875,
|
| 2121 |
+
"learning_rate": 9.898492462311558e-05,
|
| 2122 |
+
"loss": 1.3455,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.03099823524898335,
|
| 2127 |
+
"grad_norm": 0.267578125,
|
| 2128 |
+
"learning_rate": 9.897989949748744e-05,
|
| 2129 |
+
"loss": 1.3542,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.03110053965574567,
|
| 2134 |
+
"grad_norm": 0.32421875,
|
| 2135 |
+
"learning_rate": 9.89748743718593e-05,
|
| 2136 |
+
"loss": 1.6288,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.031202844062507992,
|
| 2141 |
+
"grad_norm": 0.287109375,
|
| 2142 |
+
"learning_rate": 9.896984924623117e-05,
|
| 2143 |
+
"loss": 1.567,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.031305148469270315,
|
| 2148 |
+
"grad_norm": 0.287109375,
|
| 2149 |
+
"learning_rate": 9.896482412060301e-05,
|
| 2150 |
+
"loss": 1.4221,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.03140745287603264,
|
| 2155 |
+
"grad_norm": 0.3125,
|
| 2156 |
+
"learning_rate": 9.895979899497489e-05,
|
| 2157 |
+
"loss": 1.6211,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.031509757282794953,
|
| 2162 |
+
"grad_norm": 0.279296875,
|
| 2163 |
+
"learning_rate": 9.895477386934674e-05,
|
| 2164 |
+
"loss": 1.5028,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.031612061689557276,
|
| 2169 |
+
"grad_norm": 0.353515625,
|
| 2170 |
+
"learning_rate": 9.89497487437186e-05,
|
| 2171 |
+
"loss": 1.4722,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.0317143660963196,
|
| 2176 |
+
"grad_norm": 0.29296875,
|
| 2177 |
+
"learning_rate": 9.894472361809046e-05,
|
| 2178 |
+
"loss": 1.4819,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.03181667050308192,
|
| 2183 |
+
"grad_norm": 0.427734375,
|
| 2184 |
+
"learning_rate": 9.893969849246232e-05,
|
| 2185 |
+
"loss": 1.4361,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.031918974909844244,
|
| 2190 |
+
"grad_norm": 0.27734375,
|
| 2191 |
+
"learning_rate": 9.893467336683417e-05,
|
| 2192 |
+
"loss": 1.3847,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.03202127931660656,
|
| 2197 |
+
"grad_norm": 0.30859375,
|
| 2198 |
+
"learning_rate": 9.892964824120604e-05,
|
| 2199 |
+
"loss": 1.6572,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.03212358372336888,
|
| 2204 |
+
"grad_norm": 0.27734375,
|
| 2205 |
+
"learning_rate": 9.89246231155779e-05,
|
| 2206 |
+
"loss": 1.4203,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.032225888130131206,
|
| 2211 |
+
"grad_norm": 0.291015625,
|
| 2212 |
+
"learning_rate": 9.891959798994975e-05,
|
| 2213 |
+
"loss": 1.5494,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.03232819253689353,
|
| 2218 |
+
"grad_norm": 0.27734375,
|
| 2219 |
+
"learning_rate": 9.891457286432161e-05,
|
| 2220 |
+
"loss": 1.535,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.03243049694365585,
|
| 2225 |
+
"grad_norm": 0.275390625,
|
| 2226 |
+
"learning_rate": 9.890954773869347e-05,
|
| 2227 |
+
"loss": 1.4293,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.03253280135041817,
|
| 2232 |
+
"grad_norm": 0.296875,
|
| 2233 |
+
"learning_rate": 9.890452261306533e-05,
|
| 2234 |
+
"loss": 1.5293,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.03263510575718049,
|
| 2239 |
+
"grad_norm": 0.259765625,
|
| 2240 |
+
"learning_rate": 9.889949748743718e-05,
|
| 2241 |
+
"loss": 1.321,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.03273741016394281,
|
| 2246 |
+
"grad_norm": 0.2890625,
|
| 2247 |
+
"learning_rate": 9.889447236180906e-05,
|
| 2248 |
+
"loss": 1.3901,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.032839714570705135,
|
| 2253 |
+
"grad_norm": 0.3046875,
|
| 2254 |
+
"learning_rate": 9.88894472361809e-05,
|
| 2255 |
+
"loss": 1.6575,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.03294201897746746,
|
| 2260 |
+
"grad_norm": 0.287109375,
|
| 2261 |
+
"learning_rate": 9.888442211055277e-05,
|
| 2262 |
+
"loss": 1.3796,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.03304432338422977,
|
| 2267 |
+
"grad_norm": 0.330078125,
|
| 2268 |
+
"learning_rate": 9.887939698492463e-05,
|
| 2269 |
+
"loss": 1.5548,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.033146627790992096,
|
| 2274 |
+
"grad_norm": 0.291015625,
|
| 2275 |
+
"learning_rate": 9.887437185929649e-05,
|
| 2276 |
+
"loss": 1.385,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.03324893219775442,
|
| 2281 |
+
"grad_norm": 0.283203125,
|
| 2282 |
+
"learning_rate": 9.886934673366834e-05,
|
| 2283 |
+
"loss": 1.5527,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.03335123660451674,
|
| 2288 |
+
"grad_norm": 0.302734375,
|
| 2289 |
+
"learning_rate": 9.886432160804021e-05,
|
| 2290 |
+
"loss": 1.546,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.03345354101127906,
|
| 2295 |
+
"grad_norm": 0.28515625,
|
| 2296 |
+
"learning_rate": 9.885929648241206e-05,
|
| 2297 |
+
"loss": 1.4091,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.03355584541804138,
|
| 2302 |
+
"grad_norm": 0.2890625,
|
| 2303 |
+
"learning_rate": 9.885427135678393e-05,
|
| 2304 |
+
"loss": 1.4189,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.0336581498248037,
|
| 2309 |
+
"grad_norm": 0.28515625,
|
| 2310 |
+
"learning_rate": 9.884924623115577e-05,
|
| 2311 |
+
"loss": 1.5757,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.033760454231566026,
|
| 2316 |
+
"grad_norm": 0.302734375,
|
| 2317 |
+
"learning_rate": 9.884422110552765e-05,
|
| 2318 |
+
"loss": 1.4861,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.03386275863832835,
|
| 2323 |
+
"grad_norm": 0.283203125,
|
| 2324 |
+
"learning_rate": 9.88391959798995e-05,
|
| 2325 |
+
"loss": 1.4099,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.033965063045090664,
|
| 2330 |
+
"grad_norm": 0.275390625,
|
| 2331 |
+
"learning_rate": 9.883417085427136e-05,
|
| 2332 |
+
"loss": 1.4004,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.03406736745185299,
|
| 2337 |
+
"grad_norm": 0.314453125,
|
| 2338 |
+
"learning_rate": 9.882914572864322e-05,
|
| 2339 |
+
"loss": 1.5382,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.03416967185861531,
|
| 2344 |
+
"grad_norm": 0.34765625,
|
| 2345 |
+
"learning_rate": 9.882412060301508e-05,
|
| 2346 |
+
"loss": 1.5474,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.03427197626537763,
|
| 2351 |
+
"grad_norm": 0.259765625,
|
| 2352 |
+
"learning_rate": 9.881909547738693e-05,
|
| 2353 |
+
"loss": 2.0222,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.034374280672139955,
|
| 2358 |
+
"grad_norm": 0.287109375,
|
| 2359 |
+
"learning_rate": 9.881407035175881e-05,
|
| 2360 |
+
"loss": 1.506,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.03447658507890227,
|
| 2365 |
+
"grad_norm": 0.283203125,
|
| 2366 |
+
"learning_rate": 9.880904522613066e-05,
|
| 2367 |
+
"loss": 1.5479,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.03457888948566459,
|
| 2372 |
+
"grad_norm": 0.27734375,
|
| 2373 |
+
"learning_rate": 9.880402010050252e-05,
|
| 2374 |
+
"loss": 1.4794,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.034681193892426916,
|
| 2379 |
+
"grad_norm": 0.28125,
|
| 2380 |
+
"learning_rate": 9.879899497487438e-05,
|
| 2381 |
+
"loss": 1.4704,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.03478349829918924,
|
| 2386 |
+
"grad_norm": 0.3359375,
|
| 2387 |
+
"learning_rate": 9.879396984924624e-05,
|
| 2388 |
+
"loss": 1.72,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.03488580270595156,
|
| 2393 |
+
"grad_norm": 0.259765625,
|
| 2394 |
+
"learning_rate": 9.878894472361809e-05,
|
| 2395 |
+
"loss": 1.3659,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.03498810711271388,
|
| 2400 |
+
"grad_norm": 0.28125,
|
| 2401 |
+
"learning_rate": 9.878391959798995e-05,
|
| 2402 |
+
"loss": 1.5252,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.0350904115194762,
|
| 2407 |
+
"grad_norm": 0.34765625,
|
| 2408 |
+
"learning_rate": 9.877889447236182e-05,
|
| 2409 |
+
"loss": 1.6494,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.03519271592623852,
|
| 2414 |
+
"grad_norm": 0.279296875,
|
| 2415 |
+
"learning_rate": 9.877386934673368e-05,
|
| 2416 |
+
"loss": 1.4853,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.035295020333000846,
|
| 2421 |
+
"grad_norm": 0.283203125,
|
| 2422 |
+
"learning_rate": 9.876884422110553e-05,
|
| 2423 |
+
"loss": 1.4591,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.03539732473976317,
|
| 2428 |
+
"grad_norm": 0.26953125,
|
| 2429 |
+
"learning_rate": 9.87638190954774e-05,
|
| 2430 |
+
"loss": 1.462,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.035499629146525484,
|
| 2435 |
+
"grad_norm": 0.30078125,
|
| 2436 |
+
"learning_rate": 9.875879396984925e-05,
|
| 2437 |
+
"loss": 1.5346,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.03560193355328781,
|
| 2442 |
+
"grad_norm": 0.27734375,
|
| 2443 |
+
"learning_rate": 9.875376884422111e-05,
|
| 2444 |
+
"loss": 1.4419,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.03570423796005013,
|
| 2449 |
+
"grad_norm": 0.306640625,
|
| 2450 |
+
"learning_rate": 9.874874371859297e-05,
|
| 2451 |
+
"loss": 1.5451,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.03580654236681245,
|
| 2456 |
+
"grad_norm": 0.34375,
|
| 2457 |
+
"learning_rate": 9.874371859296482e-05,
|
| 2458 |
+
"loss": 1.5113,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.035908846773574775,
|
| 2463 |
+
"grad_norm": 0.3046875,
|
| 2464 |
+
"learning_rate": 9.873869346733668e-05,
|
| 2465 |
+
"loss": 1.4416,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.03601115118033709,
|
| 2470 |
+
"grad_norm": 0.306640625,
|
| 2471 |
+
"learning_rate": 9.873366834170855e-05,
|
| 2472 |
+
"loss": 1.6185,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.03611345558709941,
|
| 2477 |
+
"grad_norm": 0.287109375,
|
| 2478 |
+
"learning_rate": 9.872864321608041e-05,
|
| 2479 |
+
"loss": 1.4294,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.036215759993861736,
|
| 2484 |
+
"grad_norm": 0.3203125,
|
| 2485 |
+
"learning_rate": 9.872361809045226e-05,
|
| 2486 |
+
"loss": 1.6404,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.03631806440062406,
|
| 2491 |
+
"grad_norm": 0.25390625,
|
| 2492 |
+
"learning_rate": 9.871859296482413e-05,
|
| 2493 |
+
"loss": 1.3696,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.03642036880738638,
|
| 2498 |
+
"grad_norm": 0.28515625,
|
| 2499 |
+
"learning_rate": 9.871356783919598e-05,
|
| 2500 |
+
"loss": 1.411,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.0365226732141487,
|
| 2505 |
+
"grad_norm": 0.337890625,
|
| 2506 |
+
"learning_rate": 9.870854271356784e-05,
|
| 2507 |
+
"loss": 1.4582,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.03662497762091102,
|
| 2512 |
+
"grad_norm": 0.41015625,
|
| 2513 |
+
"learning_rate": 9.870351758793969e-05,
|
| 2514 |
+
"loss": 1.7792,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.03672728202767334,
|
| 2519 |
+
"grad_norm": 0.33984375,
|
| 2520 |
+
"learning_rate": 9.869849246231157e-05,
|
| 2521 |
+
"loss": 1.517,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.036829586434435665,
|
| 2526 |
+
"grad_norm": 0.28125,
|
| 2527 |
+
"learning_rate": 9.869346733668342e-05,
|
| 2528 |
+
"loss": 1.3991,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.03693189084119798,
|
| 2533 |
+
"grad_norm": 0.306640625,
|
| 2534 |
+
"learning_rate": 9.868844221105528e-05,
|
| 2535 |
+
"loss": 1.5767,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.037034195247960304,
|
| 2540 |
+
"grad_norm": 0.275390625,
|
| 2541 |
+
"learning_rate": 9.868341708542714e-05,
|
| 2542 |
+
"loss": 1.4882,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.03713649965472263,
|
| 2547 |
+
"grad_norm": 0.291015625,
|
| 2548 |
+
"learning_rate": 9.8678391959799e-05,
|
| 2549 |
+
"loss": 1.5225,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.03723880406148495,
|
| 2554 |
+
"grad_norm": 0.3359375,
|
| 2555 |
+
"learning_rate": 9.867336683417085e-05,
|
| 2556 |
+
"loss": 1.4675,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.03734110846824727,
|
| 2561 |
+
"grad_norm": 0.3125,
|
| 2562 |
+
"learning_rate": 9.866834170854273e-05,
|
| 2563 |
+
"loss": 1.5146,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.03744341287500959,
|
| 2568 |
+
"grad_norm": 0.287109375,
|
| 2569 |
+
"learning_rate": 9.866331658291457e-05,
|
| 2570 |
+
"loss": 1.4159,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.03754571728177191,
|
| 2575 |
+
"grad_norm": 0.3046875,
|
| 2576 |
+
"learning_rate": 9.865829145728644e-05,
|
| 2577 |
+
"loss": 1.4227,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.03764802168853423,
|
| 2582 |
+
"grad_norm": 0.279296875,
|
| 2583 |
+
"learning_rate": 9.86532663316583e-05,
|
| 2584 |
+
"loss": 1.548,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.037750326095296556,
|
| 2589 |
+
"grad_norm": 0.27734375,
|
| 2590 |
+
"learning_rate": 9.864824120603016e-05,
|
| 2591 |
+
"loss": 1.5389,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.03785263050205888,
|
| 2596 |
+
"grad_norm": 0.283203125,
|
| 2597 |
+
"learning_rate": 9.864321608040201e-05,
|
| 2598 |
+
"loss": 1.4892,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.037954934908821195,
|
| 2603 |
+
"grad_norm": 0.30078125,
|
| 2604 |
+
"learning_rate": 9.863819095477388e-05,
|
| 2605 |
+
"loss": 1.4655,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.03805723931558352,
|
| 2610 |
+
"grad_norm": 0.2890625,
|
| 2611 |
+
"learning_rate": 9.863316582914573e-05,
|
| 2612 |
+
"loss": 1.4212,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.03815954372234584,
|
| 2617 |
+
"grad_norm": 0.298828125,
|
| 2618 |
+
"learning_rate": 9.86281407035176e-05,
|
| 2619 |
+
"loss": 1.5425,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.03826184812910816,
|
| 2624 |
+
"grad_norm": 0.2734375,
|
| 2625 |
+
"learning_rate": 9.862311557788944e-05,
|
| 2626 |
+
"loss": 1.4689,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.038364152535870485,
|
| 2631 |
+
"grad_norm": 0.279296875,
|
| 2632 |
+
"learning_rate": 9.861809045226132e-05,
|
| 2633 |
+
"loss": 1.432,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.0384664569426328,
|
| 2638 |
+
"grad_norm": 0.310546875,
|
| 2639 |
+
"learning_rate": 9.861306532663317e-05,
|
| 2640 |
+
"loss": 1.5188,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.038568761349395124,
|
| 2645 |
+
"grad_norm": 0.283203125,
|
| 2646 |
+
"learning_rate": 9.860804020100503e-05,
|
| 2647 |
+
"loss": 1.3553,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.03867106575615745,
|
| 2652 |
+
"grad_norm": 0.28515625,
|
| 2653 |
+
"learning_rate": 9.860301507537689e-05,
|
| 2654 |
+
"loss": 1.33,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.03877337016291977,
|
| 2659 |
+
"grad_norm": 0.291015625,
|
| 2660 |
+
"learning_rate": 9.859798994974875e-05,
|
| 2661 |
+
"loss": 1.5601,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.03887567456968209,
|
| 2666 |
+
"grad_norm": 0.267578125,
|
| 2667 |
+
"learning_rate": 9.85929648241206e-05,
|
| 2668 |
+
"loss": 1.4254,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.03897797897644441,
|
| 2673 |
+
"grad_norm": 0.29296875,
|
| 2674 |
+
"learning_rate": 9.858793969849246e-05,
|
| 2675 |
+
"loss": 1.5918,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.03908028338320673,
|
| 2680 |
+
"grad_norm": 0.28125,
|
| 2681 |
+
"learning_rate": 9.858291457286433e-05,
|
| 2682 |
+
"loss": 1.4162,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.03918258778996905,
|
| 2687 |
+
"grad_norm": 0.318359375,
|
| 2688 |
+
"learning_rate": 9.857788944723619e-05,
|
| 2689 |
+
"loss": 1.5642,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.039284892196731376,
|
| 2694 |
+
"grad_norm": 0.267578125,
|
| 2695 |
+
"learning_rate": 9.857286432160805e-05,
|
| 2696 |
+
"loss": 1.4412,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.0393871966034937,
|
| 2701 |
+
"grad_norm": 0.25390625,
|
| 2702 |
+
"learning_rate": 9.85678391959799e-05,
|
| 2703 |
+
"loss": 1.3841,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.039489501010256015,
|
| 2708 |
+
"grad_norm": 0.287109375,
|
| 2709 |
+
"learning_rate": 9.856281407035176e-05,
|
| 2710 |
+
"loss": 1.498,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.03959180541701834,
|
| 2715 |
+
"grad_norm": 0.27734375,
|
| 2716 |
+
"learning_rate": 9.855778894472362e-05,
|
| 2717 |
+
"loss": 1.4214,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.03969410982378066,
|
| 2722 |
+
"grad_norm": 0.294921875,
|
| 2723 |
+
"learning_rate": 9.855276381909548e-05,
|
| 2724 |
+
"loss": 1.4923,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.03979641423054298,
|
| 2729 |
+
"grad_norm": 0.3046875,
|
| 2730 |
+
"learning_rate": 9.854773869346733e-05,
|
| 2731 |
+
"loss": 1.4241,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.0398987186373053,
|
| 2736 |
+
"grad_norm": 0.302734375,
|
| 2737 |
+
"learning_rate": 9.85427135678392e-05,
|
| 2738 |
+
"loss": 1.7142,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.04000102304406762,
|
| 2743 |
+
"grad_norm": 0.3203125,
|
| 2744 |
+
"learning_rate": 9.853768844221106e-05,
|
| 2745 |
+
"loss": 1.6048,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.040103327450829944,
|
| 2750 |
+
"grad_norm": 0.27734375,
|
| 2751 |
+
"learning_rate": 9.853266331658292e-05,
|
| 2752 |
+
"loss": 1.2976,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.04020563185759227,
|
| 2757 |
+
"grad_norm": 0.27734375,
|
| 2758 |
+
"learning_rate": 9.852763819095477e-05,
|
| 2759 |
+
"loss": 1.4336,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.04030793626435459,
|
| 2764 |
+
"grad_norm": 0.306640625,
|
| 2765 |
+
"learning_rate": 9.852261306532664e-05,
|
| 2766 |
+
"loss": 1.526,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.040410240671116905,
|
| 2771 |
+
"grad_norm": 0.28125,
|
| 2772 |
+
"learning_rate": 9.851758793969849e-05,
|
| 2773 |
+
"loss": 1.5614,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.04051254507787923,
|
| 2778 |
+
"grad_norm": 0.279296875,
|
| 2779 |
+
"learning_rate": 9.851256281407035e-05,
|
| 2780 |
+
"loss": 1.5674,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.04061484948464155,
|
| 2785 |
+
"grad_norm": 0.326171875,
|
| 2786 |
+
"learning_rate": 9.850753768844222e-05,
|
| 2787 |
+
"loss": 1.4084,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.04071715389140387,
|
| 2792 |
+
"grad_norm": 0.279296875,
|
| 2793 |
+
"learning_rate": 9.850251256281408e-05,
|
| 2794 |
+
"loss": 1.4501,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.040819458298166196,
|
| 2799 |
+
"grad_norm": 0.27734375,
|
| 2800 |
+
"learning_rate": 9.849748743718593e-05,
|
| 2801 |
+
"loss": 1.5337,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.04092176270492851,
|
| 2806 |
+
"grad_norm": 0.30078125,
|
| 2807 |
+
"learning_rate": 9.84924623115578e-05,
|
| 2808 |
+
"loss": 1.4503,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.041024067111690835,
|
| 2813 |
+
"grad_norm": 0.2578125,
|
| 2814 |
+
"learning_rate": 9.848743718592965e-05,
|
| 2815 |
+
"loss": 1.44,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.04112637151845316,
|
| 2820 |
+
"grad_norm": 0.2890625,
|
| 2821 |
+
"learning_rate": 9.848241206030151e-05,
|
| 2822 |
+
"loss": 1.5709,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.04122867592521548,
|
| 2827 |
+
"grad_norm": 0.296875,
|
| 2828 |
+
"learning_rate": 9.847738693467337e-05,
|
| 2829 |
+
"loss": 1.5478,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.0413309803319778,
|
| 2834 |
+
"grad_norm": 0.2890625,
|
| 2835 |
+
"learning_rate": 9.847236180904524e-05,
|
| 2836 |
+
"loss": 1.2712,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.04143328473874012,
|
| 2841 |
+
"grad_norm": 0.29296875,
|
| 2842 |
+
"learning_rate": 9.846733668341709e-05,
|
| 2843 |
+
"loss": 1.5433,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.04153558914550244,
|
| 2848 |
+
"grad_norm": 0.341796875,
|
| 2849 |
+
"learning_rate": 9.846231155778895e-05,
|
| 2850 |
+
"loss": 1.4632,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.041637893552264764,
|
| 2855 |
+
"grad_norm": 0.2890625,
|
| 2856 |
+
"learning_rate": 9.845728643216081e-05,
|
| 2857 |
+
"loss": 1.5674,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.04174019795902709,
|
| 2862 |
+
"grad_norm": 0.29296875,
|
| 2863 |
+
"learning_rate": 9.845226130653267e-05,
|
| 2864 |
+
"loss": 1.3531,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.04184250236578941,
|
| 2869 |
+
"grad_norm": 0.380859375,
|
| 2870 |
+
"learning_rate": 9.844723618090452e-05,
|
| 2871 |
+
"loss": 1.6524,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.041944806772551725,
|
| 2876 |
+
"grad_norm": 0.322265625,
|
| 2877 |
+
"learning_rate": 9.84422110552764e-05,
|
| 2878 |
+
"loss": 1.4676,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.04204711117931405,
|
| 2883 |
+
"grad_norm": 0.2734375,
|
| 2884 |
+
"learning_rate": 9.843718592964824e-05,
|
| 2885 |
+
"loss": 1.4209,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.04214941558607637,
|
| 2890 |
+
"grad_norm": 0.333984375,
|
| 2891 |
+
"learning_rate": 9.84321608040201e-05,
|
| 2892 |
+
"loss": 1.5188,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.04225171999283869,
|
| 2897 |
+
"grad_norm": 0.3125,
|
| 2898 |
+
"learning_rate": 9.842713567839197e-05,
|
| 2899 |
+
"loss": 1.4647,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.042354024399601016,
|
| 2904 |
+
"grad_norm": 0.271484375,
|
| 2905 |
+
"learning_rate": 9.842211055276383e-05,
|
| 2906 |
+
"loss": 1.3745,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.04245632880636333,
|
| 2911 |
+
"grad_norm": 0.283203125,
|
| 2912 |
+
"learning_rate": 9.841708542713568e-05,
|
| 2913 |
+
"loss": 1.4112,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.042558633213125655,
|
| 2918 |
+
"grad_norm": 0.2470703125,
|
| 2919 |
+
"learning_rate": 9.841206030150754e-05,
|
| 2920 |
+
"loss": 1.2667,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.04266093761988798,
|
| 2925 |
+
"grad_norm": 0.279296875,
|
| 2926 |
+
"learning_rate": 9.84070351758794e-05,
|
| 2927 |
+
"loss": 1.5598,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.0427632420266503,
|
| 2932 |
+
"grad_norm": 0.291015625,
|
| 2933 |
+
"learning_rate": 9.840201005025126e-05,
|
| 2934 |
+
"loss": 1.4135,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.04286554643341262,
|
| 2939 |
+
"grad_norm": 0.296875,
|
| 2940 |
+
"learning_rate": 9.839698492462313e-05,
|
| 2941 |
+
"loss": 1.507,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.04296785084017494,
|
| 2946 |
+
"grad_norm": 0.298828125,
|
| 2947 |
+
"learning_rate": 9.839195979899497e-05,
|
| 2948 |
+
"loss": 1.4718,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.04307015524693726,
|
| 2953 |
+
"grad_norm": 0.296875,
|
| 2954 |
+
"learning_rate": 9.838693467336684e-05,
|
| 2955 |
+
"loss": 1.4224,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.043172459653699584,
|
| 2960 |
+
"grad_norm": 0.279296875,
|
| 2961 |
+
"learning_rate": 9.83819095477387e-05,
|
| 2962 |
+
"loss": 1.4815,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.04327476406046191,
|
| 2967 |
+
"grad_norm": 0.28515625,
|
| 2968 |
+
"learning_rate": 9.837688442211056e-05,
|
| 2969 |
+
"loss": 1.5333,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.04337706846722422,
|
| 2974 |
+
"grad_norm": 0.2734375,
|
| 2975 |
+
"learning_rate": 9.837185929648241e-05,
|
| 2976 |
+
"loss": 1.4247,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.043479372873986545,
|
| 2981 |
+
"grad_norm": 0.294921875,
|
| 2982 |
+
"learning_rate": 9.836683417085427e-05,
|
| 2983 |
+
"loss": 1.5218,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.04358167728074887,
|
| 2988 |
+
"grad_norm": 0.3359375,
|
| 2989 |
+
"learning_rate": 9.836180904522613e-05,
|
| 2990 |
+
"loss": 1.6057,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.04368398168751119,
|
| 2995 |
+
"grad_norm": 0.30078125,
|
| 2996 |
+
"learning_rate": 9.8356783919598e-05,
|
| 2997 |
+
"loss": 1.6459,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.04378628609427351,
|
| 3002 |
+
"grad_norm": 0.265625,
|
| 3003 |
+
"learning_rate": 9.835175879396984e-05,
|
| 3004 |
+
"loss": 1.3835,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.04388859050103583,
|
| 3009 |
+
"grad_norm": 0.322265625,
|
| 3010 |
+
"learning_rate": 9.834673366834172e-05,
|
| 3011 |
+
"loss": 1.4054,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.04399089490779815,
|
| 3016 |
+
"grad_norm": 0.251953125,
|
| 3017 |
+
"learning_rate": 9.834170854271357e-05,
|
| 3018 |
+
"loss": 1.4183,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.044093199314560474,
|
| 3023 |
+
"grad_norm": 0.29296875,
|
| 3024 |
+
"learning_rate": 9.833668341708543e-05,
|
| 3025 |
+
"loss": 1.509,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.0441955037213228,
|
| 3030 |
+
"grad_norm": 0.30859375,
|
| 3031 |
+
"learning_rate": 9.833165829145729e-05,
|
| 3032 |
+
"loss": 1.5587,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.04429780812808512,
|
| 3037 |
+
"grad_norm": 0.318359375,
|
| 3038 |
+
"learning_rate": 9.832663316582915e-05,
|
| 3039 |
+
"loss": 1.6455,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.044400112534847436,
|
| 3044 |
+
"grad_norm": 0.2578125,
|
| 3045 |
+
"learning_rate": 9.8321608040201e-05,
|
| 3046 |
+
"loss": 1.3427,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.04450241694160976,
|
| 3051 |
+
"grad_norm": 0.341796875,
|
| 3052 |
+
"learning_rate": 9.831658291457288e-05,
|
| 3053 |
+
"loss": 1.382,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.04460472134837208,
|
| 3058 |
+
"grad_norm": 0.32421875,
|
| 3059 |
+
"learning_rate": 9.831155778894473e-05,
|
| 3060 |
+
"loss": 1.42,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.044707025755134404,
|
| 3065 |
+
"grad_norm": 0.287109375,
|
| 3066 |
+
"learning_rate": 9.830653266331659e-05,
|
| 3067 |
+
"loss": 1.4817,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.04480933016189673,
|
| 3072 |
+
"grad_norm": 0.302734375,
|
| 3073 |
+
"learning_rate": 9.830150753768844e-05,
|
| 3074 |
+
"loss": 1.5843,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.04491163456865904,
|
| 3079 |
+
"grad_norm": 0.3125,
|
| 3080 |
+
"learning_rate": 9.829648241206031e-05,
|
| 3081 |
+
"loss": 1.4559,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.045013938975421365,
|
| 3086 |
+
"grad_norm": 0.25390625,
|
| 3087 |
+
"learning_rate": 9.829145728643216e-05,
|
| 3088 |
+
"loss": 1.4828,
|
| 3089 |
+
"step": 440
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.04511624338218369,
|
| 3093 |
+
"grad_norm": 0.302734375,
|
| 3094 |
+
"learning_rate": 9.828643216080402e-05,
|
| 3095 |
+
"loss": 1.7695,
|
| 3096 |
+
"step": 441
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.04521854778894601,
|
| 3100 |
+
"grad_norm": 0.26953125,
|
| 3101 |
+
"learning_rate": 9.828140703517589e-05,
|
| 3102 |
+
"loss": 1.3626,
|
| 3103 |
+
"step": 442
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.04532085219570833,
|
| 3107 |
+
"grad_norm": 0.267578125,
|
| 3108 |
+
"learning_rate": 9.827638190954775e-05,
|
| 3109 |
+
"loss": 1.3116,
|
| 3110 |
+
"step": 443
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.04542315660247065,
|
| 3114 |
+
"grad_norm": 0.263671875,
|
| 3115 |
+
"learning_rate": 9.82713567839196e-05,
|
| 3116 |
+
"loss": 1.3595,
|
| 3117 |
+
"step": 444
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.04552546100923297,
|
| 3121 |
+
"grad_norm": 0.26953125,
|
| 3122 |
+
"learning_rate": 9.826633165829147e-05,
|
| 3123 |
+
"loss": 1.5105,
|
| 3124 |
+
"step": 445
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.045627765415995294,
|
| 3128 |
+
"grad_norm": 0.26953125,
|
| 3129 |
+
"learning_rate": 9.826130653266332e-05,
|
| 3130 |
+
"loss": 1.5128,
|
| 3131 |
+
"step": 446
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.04573006982275762,
|
| 3135 |
+
"grad_norm": 0.318359375,
|
| 3136 |
+
"learning_rate": 9.825628140703518e-05,
|
| 3137 |
+
"loss": 1.3877,
|
| 3138 |
+
"step": 447
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.04583237422951994,
|
| 3142 |
+
"grad_norm": 0.283203125,
|
| 3143 |
+
"learning_rate": 9.825125628140704e-05,
|
| 3144 |
+
"loss": 1.5253,
|
| 3145 |
+
"step": 448
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.045934678636282256,
|
| 3149 |
+
"grad_norm": 0.322265625,
|
| 3150 |
+
"learning_rate": 9.82462311557789e-05,
|
| 3151 |
+
"loss": 1.3744,
|
| 3152 |
+
"step": 449
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.04603698304304458,
|
| 3156 |
+
"grad_norm": 0.279296875,
|
| 3157 |
+
"learning_rate": 9.824120603015075e-05,
|
| 3158 |
+
"loss": 1.4214,
|
| 3159 |
+
"step": 450
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.0461392874498069,
|
| 3163 |
+
"grad_norm": 0.2734375,
|
| 3164 |
+
"learning_rate": 9.823618090452262e-05,
|
| 3165 |
+
"loss": 1.3398,
|
| 3166 |
+
"step": 451
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.046241591856569224,
|
| 3170 |
+
"grad_norm": 0.279296875,
|
| 3171 |
+
"learning_rate": 9.823115577889448e-05,
|
| 3172 |
+
"loss": 1.5119,
|
| 3173 |
+
"step": 452
|
| 3174 |
+
},
|
| 3175 |
+
{
|
| 3176 |
+
"epoch": 0.04634389626333154,
|
| 3177 |
+
"grad_norm": 0.2890625,
|
| 3178 |
+
"learning_rate": 9.822613065326634e-05,
|
| 3179 |
+
"loss": 1.5934,
|
| 3180 |
+
"step": 453
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"epoch": 0.04644620067009386,
|
| 3184 |
+
"grad_norm": 0.287109375,
|
| 3185 |
+
"learning_rate": 9.822110552763819e-05,
|
| 3186 |
+
"loss": 1.4602,
|
| 3187 |
+
"step": 454
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 0.046548505076856185,
|
| 3191 |
+
"grad_norm": 0.28515625,
|
| 3192 |
+
"learning_rate": 9.821608040201005e-05,
|
| 3193 |
+
"loss": 1.5107,
|
| 3194 |
+
"step": 455
|
| 3195 |
+
},
|
| 3196 |
+
{
|
| 3197 |
+
"epoch": 0.04665080948361851,
|
| 3198 |
+
"grad_norm": 0.291015625,
|
| 3199 |
+
"learning_rate": 9.821105527638191e-05,
|
| 3200 |
+
"loss": 1.6103,
|
| 3201 |
+
"step": 456
|
| 3202 |
+
},
|
| 3203 |
+
{
|
| 3204 |
+
"epoch": 0.04675311389038083,
|
| 3205 |
+
"grad_norm": 0.3125,
|
| 3206 |
+
"learning_rate": 9.820603015075378e-05,
|
| 3207 |
+
"loss": 1.5265,
|
| 3208 |
+
"step": 457
|
| 3209 |
+
},
|
| 3210 |
+
{
|
| 3211 |
+
"epoch": 0.046855418297143146,
|
| 3212 |
+
"grad_norm": 0.255859375,
|
| 3213 |
+
"learning_rate": 9.820100502512564e-05,
|
| 3214 |
+
"loss": 1.3611,
|
| 3215 |
+
"step": 458
|
| 3216 |
+
},
|
| 3217 |
+
{
|
| 3218 |
+
"epoch": 0.04695772270390547,
|
| 3219 |
+
"grad_norm": 0.30859375,
|
| 3220 |
+
"learning_rate": 9.819597989949749e-05,
|
| 3221 |
+
"loss": 1.6182,
|
| 3222 |
+
"step": 459
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 0.04706002711066779,
|
| 3226 |
+
"grad_norm": 0.310546875,
|
| 3227 |
+
"learning_rate": 9.819095477386935e-05,
|
| 3228 |
+
"loss": 1.3604,
|
| 3229 |
+
"step": 460
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 0.047162331517430114,
|
| 3233 |
+
"grad_norm": 0.28515625,
|
| 3234 |
+
"learning_rate": 9.818592964824121e-05,
|
| 3235 |
+
"loss": 1.5613,
|
| 3236 |
+
"step": 461
|
| 3237 |
+
},
|
| 3238 |
+
{
|
| 3239 |
+
"epoch": 0.04726463592419244,
|
| 3240 |
+
"grad_norm": 0.275390625,
|
| 3241 |
+
"learning_rate": 9.818090452261307e-05,
|
| 3242 |
+
"loss": 1.5404,
|
| 3243 |
+
"step": 462
|
| 3244 |
+
},
|
| 3245 |
+
{
|
| 3246 |
+
"epoch": 0.04736694033095475,
|
| 3247 |
+
"grad_norm": 0.28125,
|
| 3248 |
+
"learning_rate": 9.817587939698492e-05,
|
| 3249 |
+
"loss": 1.5799,
|
| 3250 |
+
"step": 463
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.047469244737717076,
|
| 3254 |
+
"grad_norm": 0.255859375,
|
| 3255 |
+
"learning_rate": 9.81708542713568e-05,
|
| 3256 |
+
"loss": 1.2581,
|
| 3257 |
+
"step": 464
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.0475715491444794,
|
| 3261 |
+
"grad_norm": 0.287109375,
|
| 3262 |
+
"learning_rate": 9.816582914572864e-05,
|
| 3263 |
+
"loss": 1.4294,
|
| 3264 |
+
"step": 465
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.04767385355124172,
|
| 3268 |
+
"grad_norm": 0.28515625,
|
| 3269 |
+
"learning_rate": 9.81608040201005e-05,
|
| 3270 |
+
"loss": 1.4916,
|
| 3271 |
+
"step": 466
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.047776157958004044,
|
| 3275 |
+
"grad_norm": 0.31640625,
|
| 3276 |
+
"learning_rate": 9.815577889447236e-05,
|
| 3277 |
+
"loss": 1.3478,
|
| 3278 |
+
"step": 467
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.04787846236476636,
|
| 3282 |
+
"grad_norm": 0.28125,
|
| 3283 |
+
"learning_rate": 9.815075376884423e-05,
|
| 3284 |
+
"loss": 1.5497,
|
| 3285 |
+
"step": 468
|
| 3286 |
+
},
|
| 3287 |
+
{
|
| 3288 |
+
"epoch": 0.04798076677152868,
|
| 3289 |
+
"grad_norm": 0.287109375,
|
| 3290 |
+
"learning_rate": 9.814572864321608e-05,
|
| 3291 |
+
"loss": 1.5409,
|
| 3292 |
+
"step": 469
|
| 3293 |
+
},
|
| 3294 |
+
{
|
| 3295 |
+
"epoch": 0.048083071178291005,
|
| 3296 |
+
"grad_norm": 0.271484375,
|
| 3297 |
+
"learning_rate": 9.814070351758794e-05,
|
| 3298 |
+
"loss": 1.3988,
|
| 3299 |
+
"step": 470
|
| 3300 |
+
},
|
| 3301 |
+
{
|
| 3302 |
+
"epoch": 0.04818537558505333,
|
| 3303 |
+
"grad_norm": 0.26171875,
|
| 3304 |
+
"learning_rate": 9.81356783919598e-05,
|
| 3305 |
+
"loss": 2.4223,
|
| 3306 |
+
"step": 471
|
| 3307 |
+
},
|
| 3308 |
+
{
|
| 3309 |
+
"epoch": 0.04828767999181565,
|
| 3310 |
+
"grad_norm": 0.255859375,
|
| 3311 |
+
"learning_rate": 9.813065326633167e-05,
|
| 3312 |
+
"loss": 1.3415,
|
| 3313 |
+
"step": 472
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 0.048389984398577966,
|
| 3317 |
+
"grad_norm": 0.26953125,
|
| 3318 |
+
"learning_rate": 9.812562814070351e-05,
|
| 3319 |
+
"loss": 1.3554,
|
| 3320 |
+
"step": 473
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 0.04849228880534029,
|
| 3324 |
+
"grad_norm": 0.2734375,
|
| 3325 |
+
"learning_rate": 9.812060301507539e-05,
|
| 3326 |
+
"loss": 1.5009,
|
| 3327 |
+
"step": 474
|
| 3328 |
+
},
|
| 3329 |
+
{
|
| 3330 |
+
"epoch": 0.04859459321210261,
|
| 3331 |
+
"grad_norm": 0.3359375,
|
| 3332 |
+
"learning_rate": 9.811557788944724e-05,
|
| 3333 |
+
"loss": 1.6927,
|
| 3334 |
+
"step": 475
|
| 3335 |
+
},
|
| 3336 |
+
{
|
| 3337 |
+
"epoch": 0.048696897618864934,
|
| 3338 |
+
"grad_norm": 0.3046875,
|
| 3339 |
+
"learning_rate": 9.81105527638191e-05,
|
| 3340 |
+
"loss": 1.6283,
|
| 3341 |
+
"step": 476
|
| 3342 |
+
},
|
| 3343 |
+
{
|
| 3344 |
+
"epoch": 0.04879920202562726,
|
| 3345 |
+
"grad_norm": 0.28125,
|
| 3346 |
+
"learning_rate": 9.810552763819096e-05,
|
| 3347 |
+
"loss": 1.434,
|
| 3348 |
+
"step": 477
|
| 3349 |
+
},
|
| 3350 |
+
{
|
| 3351 |
+
"epoch": 0.04890150643238957,
|
| 3352 |
+
"grad_norm": 0.279296875,
|
| 3353 |
+
"learning_rate": 9.810050251256282e-05,
|
| 3354 |
+
"loss": 1.522,
|
| 3355 |
+
"step": 478
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 0.049003810839151896,
|
| 3359 |
+
"grad_norm": 0.287109375,
|
| 3360 |
+
"learning_rate": 9.809547738693467e-05,
|
| 3361 |
+
"loss": 1.4241,
|
| 3362 |
+
"step": 479
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 0.04910611524591422,
|
| 3366 |
+
"grad_norm": 0.27734375,
|
| 3367 |
+
"learning_rate": 9.809045226130655e-05,
|
| 3368 |
+
"loss": 1.4464,
|
| 3369 |
+
"step": 480
|
| 3370 |
+
},
|
| 3371 |
+
{
|
| 3372 |
+
"epoch": 0.04920841965267654,
|
| 3373 |
+
"grad_norm": 0.28125,
|
| 3374 |
+
"learning_rate": 9.80854271356784e-05,
|
| 3375 |
+
"loss": 1.5481,
|
| 3376 |
+
"step": 481
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 0.049310724059438864,
|
| 3380 |
+
"grad_norm": 0.322265625,
|
| 3381 |
+
"learning_rate": 9.808040201005026e-05,
|
| 3382 |
+
"loss": 1.6201,
|
| 3383 |
+
"step": 482
|
| 3384 |
+
},
|
| 3385 |
+
{
|
| 3386 |
+
"epoch": 0.04941302846620118,
|
| 3387 |
+
"grad_norm": 0.267578125,
|
| 3388 |
+
"learning_rate": 9.807537688442211e-05,
|
| 3389 |
+
"loss": 1.4617,
|
| 3390 |
+
"step": 483
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"epoch": 0.0495153328729635,
|
| 3394 |
+
"grad_norm": 0.26171875,
|
| 3395 |
+
"learning_rate": 9.807035175879398e-05,
|
| 3396 |
+
"loss": 1.5015,
|
| 3397 |
+
"step": 484
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 0.049617637279725825,
|
| 3401 |
+
"grad_norm": 0.31640625,
|
| 3402 |
+
"learning_rate": 9.806532663316583e-05,
|
| 3403 |
+
"loss": 1.5654,
|
| 3404 |
+
"step": 485
|
| 3405 |
+
},
|
| 3406 |
+
{
|
| 3407 |
+
"epoch": 0.04971994168648815,
|
| 3408 |
+
"grad_norm": 0.294921875,
|
| 3409 |
+
"learning_rate": 9.806030150753769e-05,
|
| 3410 |
+
"loss": 1.4838,
|
| 3411 |
+
"step": 486
|
| 3412 |
+
},
|
| 3413 |
+
{
|
| 3414 |
+
"epoch": 0.049822246093250463,
|
| 3415 |
+
"grad_norm": 0.283203125,
|
| 3416 |
+
"learning_rate": 9.805527638190956e-05,
|
| 3417 |
+
"loss": 1.4796,
|
| 3418 |
+
"step": 487
|
| 3419 |
+
},
|
| 3420 |
+
{
|
| 3421 |
+
"epoch": 0.049924550500012786,
|
| 3422 |
+
"grad_norm": 0.287109375,
|
| 3423 |
+
"learning_rate": 9.805025125628142e-05,
|
| 3424 |
+
"loss": 1.5272,
|
| 3425 |
+
"step": 488
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 0.05002685490677511,
|
| 3429 |
+
"grad_norm": 0.294921875,
|
| 3430 |
+
"learning_rate": 9.804522613065327e-05,
|
| 3431 |
+
"loss": 1.6112,
|
| 3432 |
+
"step": 489
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 0.05012915931353743,
|
| 3436 |
+
"grad_norm": 0.283203125,
|
| 3437 |
+
"learning_rate": 9.804020100502513e-05,
|
| 3438 |
+
"loss": 1.4579,
|
| 3439 |
+
"step": 490
|
| 3440 |
+
},
|
| 3441 |
+
{
|
| 3442 |
+
"epoch": 0.050231463720299754,
|
| 3443 |
+
"grad_norm": 0.2734375,
|
| 3444 |
+
"learning_rate": 9.803517587939699e-05,
|
| 3445 |
+
"loss": 1.4171,
|
| 3446 |
+
"step": 491
|
| 3447 |
+
},
|
| 3448 |
+
{
|
| 3449 |
+
"epoch": 0.05033376812706207,
|
| 3450 |
+
"grad_norm": 0.287109375,
|
| 3451 |
+
"learning_rate": 9.803015075376885e-05,
|
| 3452 |
+
"loss": 1.6056,
|
| 3453 |
+
"step": 492
|
| 3454 |
+
},
|
| 3455 |
+
{
|
| 3456 |
+
"epoch": 0.05043607253382439,
|
| 3457 |
+
"grad_norm": 0.27734375,
|
| 3458 |
+
"learning_rate": 9.802512562814071e-05,
|
| 3459 |
+
"loss": 1.5128,
|
| 3460 |
+
"step": 493
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 0.050538376940586716,
|
| 3464 |
+
"grad_norm": 0.28125,
|
| 3465 |
+
"learning_rate": 9.802010050251256e-05,
|
| 3466 |
+
"loss": 1.55,
|
| 3467 |
+
"step": 494
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 0.05064068134734904,
|
| 3471 |
+
"grad_norm": 0.275390625,
|
| 3472 |
+
"learning_rate": 9.801507537688442e-05,
|
| 3473 |
+
"loss": 1.5617,
|
| 3474 |
+
"step": 495
|
| 3475 |
+
},
|
| 3476 |
+
{
|
| 3477 |
+
"epoch": 0.05074298575411136,
|
| 3478 |
+
"grad_norm": 0.263671875,
|
| 3479 |
+
"learning_rate": 9.801005025125629e-05,
|
| 3480 |
+
"loss": 1.4943,
|
| 3481 |
+
"step": 496
|
| 3482 |
+
},
|
| 3483 |
+
{
|
| 3484 |
+
"epoch": 0.05084529016087368,
|
| 3485 |
+
"grad_norm": 0.271484375,
|
| 3486 |
+
"learning_rate": 9.800502512562815e-05,
|
| 3487 |
+
"loss": 1.5396,
|
| 3488 |
+
"step": 497
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 0.050947594567636,
|
| 3492 |
+
"grad_norm": 0.279296875,
|
| 3493 |
+
"learning_rate": 9.8e-05,
|
| 3494 |
+
"loss": 1.5213,
|
| 3495 |
+
"step": 498
|
| 3496 |
+
},
|
| 3497 |
+
{
|
| 3498 |
+
"epoch": 0.05104989897439832,
|
| 3499 |
+
"grad_norm": 0.26953125,
|
| 3500 |
+
"learning_rate": 9.799497487437186e-05,
|
| 3501 |
+
"loss": 1.3655,
|
| 3502 |
+
"step": 499
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.051152203381160645,
|
| 3506 |
+
"grad_norm": 0.302734375,
|
| 3507 |
+
"learning_rate": 9.798994974874372e-05,
|
| 3508 |
+
"loss": 1.4502,
|
| 3509 |
+
"step": 500
|
| 3510 |
+
}
|
| 3511 |
+
],
|
| 3512 |
+
"logging_steps": 1,
|
| 3513 |
+
"max_steps": 20000,
|
| 3514 |
+
"num_input_tokens_seen": 0,
|
| 3515 |
+
"num_train_epochs": 3,
|
| 3516 |
+
"save_steps": 500,
|
| 3517 |
+
"stateful_callbacks": {
|
| 3518 |
+
"TrainerControl": {
|
| 3519 |
+
"args": {
|
| 3520 |
+
"should_epoch_stop": false,
|
| 3521 |
+
"should_evaluate": false,
|
| 3522 |
+
"should_log": false,
|
| 3523 |
+
"should_save": true,
|
| 3524 |
+
"should_training_stop": false
|
| 3525 |
+
},
|
| 3526 |
+
"attributes": {}
|
| 3527 |
+
}
|
| 3528 |
+
},
|
| 3529 |
+
"total_flos": 1.29157249892352e+17,
|
| 3530 |
+
"train_batch_size": 2,
|
| 3531 |
+
"trial_name": null,
|
| 3532 |
+
"trial_params": null
|
| 3533 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c7cc2b8534cbe11a9f9621d87c771e43fa663b9157b0d214309fec66819af42
|
| 3 |
+
size 5560
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|