Upload model
Browse files- config.json +40 -0
- generation_config.json +8 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +26 -0
- trainer_state.json +1947 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/madaraszg/mamba/test_mamba_wsd/checkpoint-570000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MambaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"conv_kernel": 4,
|
| 8 |
+
"d_inner": 1536,
|
| 9 |
+
"d_model": 768,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"expand": 2,
|
| 12 |
+
"fused_add_norm": true,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"initializer_range": 0.1,
|
| 16 |
+
"intermediate_size": 1536,
|
| 17 |
+
"layer_norm_epsilon": 1e-05,
|
| 18 |
+
"model_type": "mamba",
|
| 19 |
+
"n_layer": 24,
|
| 20 |
+
"num_hidden_layers": 24,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"pad_vocab_size_multiple": 8,
|
| 23 |
+
"rescale_prenorm_residual": false,
|
| 24 |
+
"residual_in_fp32": true,
|
| 25 |
+
"rms_norm": true,
|
| 26 |
+
"ssm_cfg": {},
|
| 27 |
+
"state_size": 16,
|
| 28 |
+
"time_step_floor": 0.0001,
|
| 29 |
+
"time_step_init_scheme": "random",
|
| 30 |
+
"time_step_max": 0.1,
|
| 31 |
+
"time_step_min": 0.001,
|
| 32 |
+
"time_step_rank": 48,
|
| 33 |
+
"time_step_scale": 1.0,
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.41.2",
|
| 36 |
+
"use_bias": false,
|
| 37 |
+
"use_cache": false,
|
| 38 |
+
"use_conv_bias": true,
|
| 39 |
+
"vocab_size": 52000
|
| 40 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.41.2",
|
| 7 |
+
"use_cache": false
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a311dc3bee49f979d95d6c130e6b6b4df18d8a6b9fca0ce3a9c6e17a08c9061
|
| 3 |
+
size 521905938
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|endoftext|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<|endoftext|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": false,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": true,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
}
|
| 14 |
+
},
|
| 15 |
+
"bos_token": "<|endoftext|>",
|
| 16 |
+
"clean_up_tokenization_spaces": true,
|
| 17 |
+
"eos_token": "<|endoftext|>",
|
| 18 |
+
"max_length": 2048,
|
| 19 |
+
"model_max_length": 1024,
|
| 20 |
+
"pad_token": "<|endoftext|>",
|
| 21 |
+
"stride": 0,
|
| 22 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 23 |
+
"truncation_side": "right",
|
| 24 |
+
"truncation_strategy": "longest_first",
|
| 25 |
+
"unk_token": "<|endoftext|>"
|
| 26 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1947 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9786327442581368,
|
| 5 |
+
"eval_steps": 15000,
|
| 6 |
+
"global_step": 370000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.003967430044289744,
|
| 13 |
+
"grad_norm": 1.5698450803756714,
|
| 14 |
+
"learning_rate": 2e-05,
|
| 15 |
+
"loss": 4.231,
|
| 16 |
+
"step": 1500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007934860088579488,
|
| 20 |
+
"grad_norm": 1.8081231117248535,
|
| 21 |
+
"learning_rate": 2e-05,
|
| 22 |
+
"loss": 4.1347,
|
| 23 |
+
"step": 3000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011902290132869232,
|
| 27 |
+
"grad_norm": 1.1643619537353516,
|
| 28 |
+
"learning_rate": 2e-05,
|
| 29 |
+
"loss": 4.1056,
|
| 30 |
+
"step": 4500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.015869720177158975,
|
| 34 |
+
"grad_norm": 1.3358386754989624,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 4.0762,
|
| 37 |
+
"step": 6000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.019837150221448722,
|
| 41 |
+
"grad_norm": 1.3741443157196045,
|
| 42 |
+
"learning_rate": 2e-05,
|
| 43 |
+
"loss": 4.0576,
|
| 44 |
+
"step": 7500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.023804580265738465,
|
| 48 |
+
"grad_norm": 1.3668568134307861,
|
| 49 |
+
"learning_rate": 2e-05,
|
| 50 |
+
"loss": 4.0384,
|
| 51 |
+
"step": 9000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.027772010310028208,
|
| 55 |
+
"grad_norm": 1.714483618736267,
|
| 56 |
+
"learning_rate": 2e-05,
|
| 57 |
+
"loss": 4.0187,
|
| 58 |
+
"step": 10500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03173944035431795,
|
| 62 |
+
"grad_norm": 1.3867794275283813,
|
| 63 |
+
"learning_rate": 2e-05,
|
| 64 |
+
"loss": 4.0159,
|
| 65 |
+
"step": 12000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0357068703986077,
|
| 69 |
+
"grad_norm": 1.603532075881958,
|
| 70 |
+
"learning_rate": 2e-05,
|
| 71 |
+
"loss": 3.9967,
|
| 72 |
+
"step": 13500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.039674300442897444,
|
| 76 |
+
"grad_norm": 1.5365407466888428,
|
| 77 |
+
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 3.9848,
|
| 79 |
+
"step": 15000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.039674300442897444,
|
| 83 |
+
"eval_loss": 3.973594903945923,
|
| 84 |
+
"eval_runtime": 509.3826,
|
| 85 |
+
"eval_samples_per_second": 131.161,
|
| 86 |
+
"eval_steps_per_second": 32.791,
|
| 87 |
+
"step": 15000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04364173048718718,
|
| 91 |
+
"grad_norm": 1.728336215019226,
|
| 92 |
+
"learning_rate": 2e-05,
|
| 93 |
+
"loss": 3.9812,
|
| 94 |
+
"step": 16500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.04760916053147693,
|
| 98 |
+
"grad_norm": 1.8370418548583984,
|
| 99 |
+
"learning_rate": 2e-05,
|
| 100 |
+
"loss": 3.9633,
|
| 101 |
+
"step": 18000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.051576590575766676,
|
| 105 |
+
"grad_norm": 1.6304991245269775,
|
| 106 |
+
"learning_rate": 2e-05,
|
| 107 |
+
"loss": 3.9549,
|
| 108 |
+
"step": 19500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.055544020620056415,
|
| 112 |
+
"grad_norm": 1.4472306966781616,
|
| 113 |
+
"learning_rate": 2e-05,
|
| 114 |
+
"loss": 3.9518,
|
| 115 |
+
"step": 21000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05951145066434616,
|
| 119 |
+
"grad_norm": 2.202906370162964,
|
| 120 |
+
"learning_rate": 2e-05,
|
| 121 |
+
"loss": 3.9497,
|
| 122 |
+
"step": 22500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.0634788807086359,
|
| 126 |
+
"grad_norm": 1.5186316967010498,
|
| 127 |
+
"learning_rate": 2e-05,
|
| 128 |
+
"loss": 3.9362,
|
| 129 |
+
"step": 24000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06744631075292565,
|
| 133 |
+
"grad_norm": 2.0651211738586426,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 3.932,
|
| 136 |
+
"step": 25500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0714137407972154,
|
| 140 |
+
"grad_norm": 1.750518560409546,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 3.9276,
|
| 143 |
+
"step": 27000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07538117084150514,
|
| 147 |
+
"grad_norm": 2.3646767139434814,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 3.9108,
|
| 150 |
+
"step": 28500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.07934860088579489,
|
| 154 |
+
"grad_norm": 1.3822572231292725,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 3.9242,
|
| 157 |
+
"step": 30000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.07934860088579489,
|
| 161 |
+
"eval_loss": 3.8984525203704834,
|
| 162 |
+
"eval_runtime": 509.4664,
|
| 163 |
+
"eval_samples_per_second": 131.139,
|
| 164 |
+
"eval_steps_per_second": 32.785,
|
| 165 |
+
"step": 30000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.08331603093008462,
|
| 169 |
+
"grad_norm": 1.394733190536499,
|
| 170 |
+
"learning_rate": 2e-05,
|
| 171 |
+
"loss": 3.9112,
|
| 172 |
+
"step": 31500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.08728346097437437,
|
| 176 |
+
"grad_norm": 2.038745641708374,
|
| 177 |
+
"learning_rate": 2e-05,
|
| 178 |
+
"loss": 3.9065,
|
| 179 |
+
"step": 33000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.09125089101866411,
|
| 183 |
+
"grad_norm": 1.3629461526870728,
|
| 184 |
+
"learning_rate": 2e-05,
|
| 185 |
+
"loss": 3.8944,
|
| 186 |
+
"step": 34500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.09521832106295386,
|
| 190 |
+
"grad_norm": 1.59019136428833,
|
| 191 |
+
"learning_rate": 2e-05,
|
| 192 |
+
"loss": 3.899,
|
| 193 |
+
"step": 36000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.0991857511072436,
|
| 197 |
+
"grad_norm": 1.7221766710281372,
|
| 198 |
+
"learning_rate": 2e-05,
|
| 199 |
+
"loss": 3.8944,
|
| 200 |
+
"step": 37500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.10315318115153335,
|
| 204 |
+
"grad_norm": 1.5057470798492432,
|
| 205 |
+
"learning_rate": 2e-05,
|
| 206 |
+
"loss": 3.8929,
|
| 207 |
+
"step": 39000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.10712061119582308,
|
| 211 |
+
"grad_norm": 1.812050223350525,
|
| 212 |
+
"learning_rate": 2e-05,
|
| 213 |
+
"loss": 3.8788,
|
| 214 |
+
"step": 40500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.11108804124011283,
|
| 218 |
+
"grad_norm": 1.7102704048156738,
|
| 219 |
+
"learning_rate": 2e-05,
|
| 220 |
+
"loss": 3.8753,
|
| 221 |
+
"step": 42000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.11505547128440258,
|
| 225 |
+
"grad_norm": 1.408778190612793,
|
| 226 |
+
"learning_rate": 2e-05,
|
| 227 |
+
"loss": 3.8705,
|
| 228 |
+
"step": 43500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.11902290132869232,
|
| 232 |
+
"grad_norm": 1.8378455638885498,
|
| 233 |
+
"learning_rate": 2e-05,
|
| 234 |
+
"loss": 3.8678,
|
| 235 |
+
"step": 45000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.11902290132869232,
|
| 239 |
+
"eval_loss": 3.851247787475586,
|
| 240 |
+
"eval_runtime": 509.1413,
|
| 241 |
+
"eval_samples_per_second": 131.223,
|
| 242 |
+
"eval_steps_per_second": 32.806,
|
| 243 |
+
"step": 45000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.12299033137298207,
|
| 247 |
+
"grad_norm": 1.7914859056472778,
|
| 248 |
+
"learning_rate": 2e-05,
|
| 249 |
+
"loss": 3.864,
|
| 250 |
+
"step": 46500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.1269577614172718,
|
| 254 |
+
"grad_norm": 1.6122477054595947,
|
| 255 |
+
"learning_rate": 2e-05,
|
| 256 |
+
"loss": 3.8561,
|
| 257 |
+
"step": 48000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.13092519146156156,
|
| 261 |
+
"grad_norm": 1.827723741531372,
|
| 262 |
+
"learning_rate": 2e-05,
|
| 263 |
+
"loss": 3.8639,
|
| 264 |
+
"step": 49500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.1348926215058513,
|
| 268 |
+
"grad_norm": 1.9765942096710205,
|
| 269 |
+
"learning_rate": 2e-05,
|
| 270 |
+
"loss": 3.8486,
|
| 271 |
+
"step": 51000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.13886005155014106,
|
| 275 |
+
"grad_norm": 1.3416199684143066,
|
| 276 |
+
"learning_rate": 2e-05,
|
| 277 |
+
"loss": 3.8514,
|
| 278 |
+
"step": 52500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.1428274815944308,
|
| 282 |
+
"grad_norm": 2.1308629512786865,
|
| 283 |
+
"learning_rate": 2e-05,
|
| 284 |
+
"loss": 3.834,
|
| 285 |
+
"step": 54000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.14679491163872052,
|
| 289 |
+
"grad_norm": 1.630344271659851,
|
| 290 |
+
"learning_rate": 2e-05,
|
| 291 |
+
"loss": 3.8438,
|
| 292 |
+
"step": 55500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.15076234168301028,
|
| 296 |
+
"grad_norm": 1.311733603477478,
|
| 297 |
+
"learning_rate": 2e-05,
|
| 298 |
+
"loss": 3.8387,
|
| 299 |
+
"step": 57000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.15472977172730001,
|
| 303 |
+
"grad_norm": 1.2008074522018433,
|
| 304 |
+
"learning_rate": 2e-05,
|
| 305 |
+
"loss": 3.831,
|
| 306 |
+
"step": 58500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.15869720177158977,
|
| 310 |
+
"grad_norm": 1.804685115814209,
|
| 311 |
+
"learning_rate": 2e-05,
|
| 312 |
+
"loss": 3.8369,
|
| 313 |
+
"step": 60000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.15869720177158977,
|
| 317 |
+
"eval_loss": 3.8154687881469727,
|
| 318 |
+
"eval_runtime": 509.3502,
|
| 319 |
+
"eval_samples_per_second": 131.169,
|
| 320 |
+
"eval_steps_per_second": 32.793,
|
| 321 |
+
"step": 60000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.1626646318158795,
|
| 325 |
+
"grad_norm": 2.4677481651306152,
|
| 326 |
+
"learning_rate": 2e-05,
|
| 327 |
+
"loss": 3.8405,
|
| 328 |
+
"step": 61500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.16663206186016924,
|
| 332 |
+
"grad_norm": 3.788940668106079,
|
| 333 |
+
"learning_rate": 2e-05,
|
| 334 |
+
"loss": 3.8371,
|
| 335 |
+
"step": 63000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.170599491904459,
|
| 339 |
+
"grad_norm": 1.9231897592544556,
|
| 340 |
+
"learning_rate": 2e-05,
|
| 341 |
+
"loss": 3.8197,
|
| 342 |
+
"step": 64500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.17456692194874873,
|
| 346 |
+
"grad_norm": 2.9989049434661865,
|
| 347 |
+
"learning_rate": 2e-05,
|
| 348 |
+
"loss": 3.8248,
|
| 349 |
+
"step": 66000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.1785343519930385,
|
| 353 |
+
"grad_norm": 1.5718803405761719,
|
| 354 |
+
"learning_rate": 2e-05,
|
| 355 |
+
"loss": 3.818,
|
| 356 |
+
"step": 67500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.18250178203732823,
|
| 360 |
+
"grad_norm": 2.03694748878479,
|
| 361 |
+
"learning_rate": 2e-05,
|
| 362 |
+
"loss": 3.8277,
|
| 363 |
+
"step": 69000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.18646921208161796,
|
| 367 |
+
"grad_norm": 3.6177451610565186,
|
| 368 |
+
"learning_rate": 2e-05,
|
| 369 |
+
"loss": 3.8079,
|
| 370 |
+
"step": 70500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.19043664212590772,
|
| 374 |
+
"grad_norm": 1.6187258958816528,
|
| 375 |
+
"learning_rate": 2e-05,
|
| 376 |
+
"loss": 3.8093,
|
| 377 |
+
"step": 72000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.19440407217019745,
|
| 381 |
+
"grad_norm": 1.4624522924423218,
|
| 382 |
+
"learning_rate": 2e-05,
|
| 383 |
+
"loss": 3.8206,
|
| 384 |
+
"step": 73500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.1983715022144872,
|
| 388 |
+
"grad_norm": 1.632415533065796,
|
| 389 |
+
"learning_rate": 2e-05,
|
| 390 |
+
"loss": 3.8037,
|
| 391 |
+
"step": 75000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.1983715022144872,
|
| 395 |
+
"eval_loss": 3.787764072418213,
|
| 396 |
+
"eval_runtime": 515.1464,
|
| 397 |
+
"eval_samples_per_second": 129.693,
|
| 398 |
+
"eval_steps_per_second": 32.424,
|
| 399 |
+
"step": 75000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.20233893225877694,
|
| 403 |
+
"grad_norm": 1.420238971710205,
|
| 404 |
+
"learning_rate": 2e-05,
|
| 405 |
+
"loss": 3.8085,
|
| 406 |
+
"step": 76500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.2063063623030667,
|
| 410 |
+
"grad_norm": 1.5415395498275757,
|
| 411 |
+
"learning_rate": 2e-05,
|
| 412 |
+
"loss": 3.8071,
|
| 413 |
+
"step": 78000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.21027379234735644,
|
| 417 |
+
"grad_norm": 1.9765852689743042,
|
| 418 |
+
"learning_rate": 2e-05,
|
| 419 |
+
"loss": 3.797,
|
| 420 |
+
"step": 79500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.21424122239164617,
|
| 424 |
+
"grad_norm": 2.0840232372283936,
|
| 425 |
+
"learning_rate": 2e-05,
|
| 426 |
+
"loss": 3.7955,
|
| 427 |
+
"step": 81000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.21820865243593593,
|
| 431 |
+
"grad_norm": 1.6770437955856323,
|
| 432 |
+
"learning_rate": 2e-05,
|
| 433 |
+
"loss": 3.7961,
|
| 434 |
+
"step": 82500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.22217608248022566,
|
| 438 |
+
"grad_norm": 2.207707166671753,
|
| 439 |
+
"learning_rate": 2e-05,
|
| 440 |
+
"loss": 3.7832,
|
| 441 |
+
"step": 84000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.22614351252451542,
|
| 445 |
+
"grad_norm": 1.7404212951660156,
|
| 446 |
+
"learning_rate": 2e-05,
|
| 447 |
+
"loss": 3.7895,
|
| 448 |
+
"step": 85500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.23011094256880515,
|
| 452 |
+
"grad_norm": 1.492181658744812,
|
| 453 |
+
"learning_rate": 2e-05,
|
| 454 |
+
"loss": 3.7876,
|
| 455 |
+
"step": 87000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.2340783726130949,
|
| 459 |
+
"grad_norm": 1.1175812482833862,
|
| 460 |
+
"learning_rate": 2e-05,
|
| 461 |
+
"loss": 3.792,
|
| 462 |
+
"step": 88500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.23804580265738465,
|
| 466 |
+
"grad_norm": 1.816019892692566,
|
| 467 |
+
"learning_rate": 2e-05,
|
| 468 |
+
"loss": 3.7788,
|
| 469 |
+
"step": 90000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.23804580265738465,
|
| 473 |
+
"eval_loss": 3.7638604640960693,
|
| 474 |
+
"eval_runtime": 517.2086,
|
| 475 |
+
"eval_samples_per_second": 129.176,
|
| 476 |
+
"eval_steps_per_second": 32.295,
|
| 477 |
+
"step": 90000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.24201323270167438,
|
| 481 |
+
"grad_norm": 2.108680009841919,
|
| 482 |
+
"learning_rate": 2e-05,
|
| 483 |
+
"loss": 3.7833,
|
| 484 |
+
"step": 91500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.24598066274596414,
|
| 488 |
+
"grad_norm": 1.6215394735336304,
|
| 489 |
+
"learning_rate": 2e-05,
|
| 490 |
+
"loss": 3.7684,
|
| 491 |
+
"step": 93000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.24994809279025387,
|
| 495 |
+
"grad_norm": 1.3144052028656006,
|
| 496 |
+
"learning_rate": 2e-05,
|
| 497 |
+
"loss": 3.7716,
|
| 498 |
+
"step": 94500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.2539155228345436,
|
| 502 |
+
"grad_norm": 1.788275957107544,
|
| 503 |
+
"learning_rate": 2e-05,
|
| 504 |
+
"loss": 3.7698,
|
| 505 |
+
"step": 96000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.25788295287883334,
|
| 509 |
+
"grad_norm": 2.0639145374298096,
|
| 510 |
+
"learning_rate": 2e-05,
|
| 511 |
+
"loss": 3.7684,
|
| 512 |
+
"step": 97500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.2618503829231231,
|
| 516 |
+
"grad_norm": 1.38687002658844,
|
| 517 |
+
"learning_rate": 2e-05,
|
| 518 |
+
"loss": 3.7696,
|
| 519 |
+
"step": 99000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.26581781296741286,
|
| 523 |
+
"grad_norm": 1.4817004203796387,
|
| 524 |
+
"learning_rate": 2e-05,
|
| 525 |
+
"loss": 3.7673,
|
| 526 |
+
"step": 100500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.2697852430117026,
|
| 530 |
+
"grad_norm": 1.7259806394577026,
|
| 531 |
+
"learning_rate": 2e-05,
|
| 532 |
+
"loss": 3.7602,
|
| 533 |
+
"step": 102000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.2737526730559923,
|
| 537 |
+
"grad_norm": 1.340493083000183,
|
| 538 |
+
"learning_rate": 2e-05,
|
| 539 |
+
"loss": 3.7514,
|
| 540 |
+
"step": 103500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.2777201031002821,
|
| 544 |
+
"grad_norm": 2.0217068195343018,
|
| 545 |
+
"learning_rate": 2e-05,
|
| 546 |
+
"loss": 3.7618,
|
| 547 |
+
"step": 105000
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.2777201031002821,
|
| 551 |
+
"eval_loss": 3.743399143218994,
|
| 552 |
+
"eval_runtime": 527.3734,
|
| 553 |
+
"eval_samples_per_second": 126.686,
|
| 554 |
+
"eval_steps_per_second": 31.672,
|
| 555 |
+
"step": 105000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.28168753314457184,
|
| 559 |
+
"grad_norm": 1.1739765405654907,
|
| 560 |
+
"learning_rate": 2e-05,
|
| 561 |
+
"loss": 3.7599,
|
| 562 |
+
"step": 106500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2856549631888616,
|
| 566 |
+
"grad_norm": 1.692695140838623,
|
| 567 |
+
"learning_rate": 2e-05,
|
| 568 |
+
"loss": 3.7447,
|
| 569 |
+
"step": 108000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2896223932331513,
|
| 573 |
+
"grad_norm": 1.8593926429748535,
|
| 574 |
+
"learning_rate": 2e-05,
|
| 575 |
+
"loss": 3.7533,
|
| 576 |
+
"step": 109500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29358982327744104,
|
| 580 |
+
"grad_norm": 2.278003454208374,
|
| 581 |
+
"learning_rate": 2e-05,
|
| 582 |
+
"loss": 3.7555,
|
| 583 |
+
"step": 111000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.29755725332173083,
|
| 587 |
+
"grad_norm": 1.2629317045211792,
|
| 588 |
+
"learning_rate": 2e-05,
|
| 589 |
+
"loss": 3.7502,
|
| 590 |
+
"step": 112500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.30152468336602056,
|
| 594 |
+
"grad_norm": 1.8427543640136719,
|
| 595 |
+
"learning_rate": 2e-05,
|
| 596 |
+
"loss": 3.7669,
|
| 597 |
+
"step": 114000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3054921134103103,
|
| 601 |
+
"grad_norm": 2.1399271488189697,
|
| 602 |
+
"learning_rate": 2e-05,
|
| 603 |
+
"loss": 3.7441,
|
| 604 |
+
"step": 115500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.30945954345460003,
|
| 608 |
+
"grad_norm": 1.726097822189331,
|
| 609 |
+
"learning_rate": 2e-05,
|
| 610 |
+
"loss": 3.7413,
|
| 611 |
+
"step": 117000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.31342697349888976,
|
| 615 |
+
"grad_norm": 2.9889960289001465,
|
| 616 |
+
"learning_rate": 2e-05,
|
| 617 |
+
"loss": 3.7435,
|
| 618 |
+
"step": 118500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.31739440354317955,
|
| 622 |
+
"grad_norm": 1.6429541110992432,
|
| 623 |
+
"learning_rate": 2e-05,
|
| 624 |
+
"loss": 3.7321,
|
| 625 |
+
"step": 120000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.31739440354317955,
|
| 629 |
+
"eval_loss": 3.725595235824585,
|
| 630 |
+
"eval_runtime": 507.8039,
|
| 631 |
+
"eval_samples_per_second": 131.568,
|
| 632 |
+
"eval_steps_per_second": 32.893,
|
| 633 |
+
"step": 120000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.3213618335874693,
|
| 637 |
+
"grad_norm": 1.523931622505188,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 3.7449,
|
| 640 |
+
"step": 121500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.325329263631759,
|
| 644 |
+
"grad_norm": 1.1351861953735352,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 3.74,
|
| 647 |
+
"step": 123000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.32929669367604875,
|
| 651 |
+
"grad_norm": 1.9448051452636719,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 3.738,
|
| 654 |
+
"step": 124500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3332641237203385,
|
| 658 |
+
"grad_norm": 2.2584104537963867,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 3.7349,
|
| 661 |
+
"step": 126000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.33723155376462827,
|
| 665 |
+
"grad_norm": 1.9008771181106567,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 3.7294,
|
| 668 |
+
"step": 127500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.341198983808918,
|
| 672 |
+
"grad_norm": 1.5307859182357788,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 3.7374,
|
| 675 |
+
"step": 129000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.34516641385320773,
|
| 679 |
+
"grad_norm": 1.4004909992218018,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 3.739,
|
| 682 |
+
"step": 130500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.34913384389749746,
|
| 686 |
+
"grad_norm": 2.54671573638916,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 3.7185,
|
| 689 |
+
"step": 132000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.3531012739417872,
|
| 693 |
+
"grad_norm": 1.5276941061019897,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 3.7359,
|
| 696 |
+
"step": 133500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.357068703986077,
|
| 700 |
+
"grad_norm": 1.8586397171020508,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 3.7275,
|
| 703 |
+
"step": 135000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.357068703986077,
|
| 707 |
+
"eval_loss": 3.708439826965332,
|
| 708 |
+
"eval_runtime": 508.0774,
|
| 709 |
+
"eval_samples_per_second": 131.498,
|
| 710 |
+
"eval_steps_per_second": 32.875,
|
| 711 |
+
"step": 135000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.3610361340303667,
|
| 715 |
+
"grad_norm": 1.7233084440231323,
|
| 716 |
+
"learning_rate": 2e-05,
|
| 717 |
+
"loss": 3.7294,
|
| 718 |
+
"step": 136500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.36500356407465645,
|
| 722 |
+
"grad_norm": 1.4811041355133057,
|
| 723 |
+
"learning_rate": 2e-05,
|
| 724 |
+
"loss": 3.7377,
|
| 725 |
+
"step": 138000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.3689709941189462,
|
| 729 |
+
"grad_norm": 1.778673768043518,
|
| 730 |
+
"learning_rate": 2e-05,
|
| 731 |
+
"loss": 3.7252,
|
| 732 |
+
"step": 139500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.3729384241632359,
|
| 736 |
+
"grad_norm": 1.6126329898834229,
|
| 737 |
+
"learning_rate": 2e-05,
|
| 738 |
+
"loss": 3.7139,
|
| 739 |
+
"step": 141000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.3769058542075257,
|
| 743 |
+
"grad_norm": 1.7368344068527222,
|
| 744 |
+
"learning_rate": 2e-05,
|
| 745 |
+
"loss": 3.7165,
|
| 746 |
+
"step": 142500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.38087328425181544,
|
| 750 |
+
"grad_norm": 1.5401374101638794,
|
| 751 |
+
"learning_rate": 2e-05,
|
| 752 |
+
"loss": 3.7141,
|
| 753 |
+
"step": 144000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.38484071429610517,
|
| 757 |
+
"grad_norm": 1.4228651523590088,
|
| 758 |
+
"learning_rate": 2e-05,
|
| 759 |
+
"loss": 3.7132,
|
| 760 |
+
"step": 145500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.3888081443403949,
|
| 764 |
+
"grad_norm": 1.5804122686386108,
|
| 765 |
+
"learning_rate": 2e-05,
|
| 766 |
+
"loss": 3.7199,
|
| 767 |
+
"step": 147000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.3927755743846847,
|
| 771 |
+
"grad_norm": 1.7531423568725586,
|
| 772 |
+
"learning_rate": 2e-05,
|
| 773 |
+
"loss": 3.7251,
|
| 774 |
+
"step": 148500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.3967430044289744,
|
| 778 |
+
"grad_norm": 2.62064528465271,
|
| 779 |
+
"learning_rate": 2e-05,
|
| 780 |
+
"loss": 3.7075,
|
| 781 |
+
"step": 150000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.3967430044289744,
|
| 785 |
+
"eval_loss": 3.693021535873413,
|
| 786 |
+
"eval_runtime": 508.4788,
|
| 787 |
+
"eval_samples_per_second": 131.394,
|
| 788 |
+
"eval_steps_per_second": 32.849,
|
| 789 |
+
"step": 150000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.40071043447326415,
|
| 793 |
+
"grad_norm": 2.2958357334136963,
|
| 794 |
+
"learning_rate": 2e-05,
|
| 795 |
+
"loss": 3.7145,
|
| 796 |
+
"step": 151500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.4046778645175539,
|
| 800 |
+
"grad_norm": 1.2222336530685425,
|
| 801 |
+
"learning_rate": 2e-05,
|
| 802 |
+
"loss": 3.7229,
|
| 803 |
+
"step": 153000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.4086452945618436,
|
| 807 |
+
"grad_norm": 1.8726732730865479,
|
| 808 |
+
"learning_rate": 2e-05,
|
| 809 |
+
"loss": 3.7075,
|
| 810 |
+
"step": 154500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.4126127246061334,
|
| 814 |
+
"grad_norm": 2.425828218460083,
|
| 815 |
+
"learning_rate": 2e-05,
|
| 816 |
+
"loss": 3.7002,
|
| 817 |
+
"step": 156000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.41658015465042314,
|
| 821 |
+
"grad_norm": 1.6482657194137573,
|
| 822 |
+
"learning_rate": 2e-05,
|
| 823 |
+
"loss": 3.7024,
|
| 824 |
+
"step": 157500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.4205475846947129,
|
| 828 |
+
"grad_norm": 2.0351409912109375,
|
| 829 |
+
"learning_rate": 2e-05,
|
| 830 |
+
"loss": 3.7079,
|
| 831 |
+
"step": 159000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.4245150147390026,
|
| 835 |
+
"grad_norm": 1.5541174411773682,
|
| 836 |
+
"learning_rate": 2e-05,
|
| 837 |
+
"loss": 3.6976,
|
| 838 |
+
"step": 160500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.42848244478329234,
|
| 842 |
+
"grad_norm": 2.035759687423706,
|
| 843 |
+
"learning_rate": 2e-05,
|
| 844 |
+
"loss": 3.6955,
|
| 845 |
+
"step": 162000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.4324498748275821,
|
| 849 |
+
"grad_norm": 2.4348483085632324,
|
| 850 |
+
"learning_rate": 2e-05,
|
| 851 |
+
"loss": 3.6893,
|
| 852 |
+
"step": 163500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.43641730487187186,
|
| 856 |
+
"grad_norm": 2.0152392387390137,
|
| 857 |
+
"learning_rate": 2e-05,
|
| 858 |
+
"loss": 3.6942,
|
| 859 |
+
"step": 165000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.43641730487187186,
|
| 863 |
+
"eval_loss": 3.679271697998047,
|
| 864 |
+
"eval_runtime": 508.381,
|
| 865 |
+
"eval_samples_per_second": 131.419,
|
| 866 |
+
"eval_steps_per_second": 32.855,
|
| 867 |
+
"step": 165000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.4403847349161616,
|
| 871 |
+
"grad_norm": 2.1153664588928223,
|
| 872 |
+
"learning_rate": 2e-05,
|
| 873 |
+
"loss": 3.6959,
|
| 874 |
+
"step": 166500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.4443521649604513,
|
| 878 |
+
"grad_norm": 1.2098298072814941,
|
| 879 |
+
"learning_rate": 2e-05,
|
| 880 |
+
"loss": 3.6839,
|
| 881 |
+
"step": 168000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.44831959500474106,
|
| 885 |
+
"grad_norm": 2.01379132270813,
|
| 886 |
+
"learning_rate": 2e-05,
|
| 887 |
+
"loss": 3.687,
|
| 888 |
+
"step": 169500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.45228702504903084,
|
| 892 |
+
"grad_norm": 1.823215365409851,
|
| 893 |
+
"learning_rate": 2e-05,
|
| 894 |
+
"loss": 3.6932,
|
| 895 |
+
"step": 171000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.4562544550933206,
|
| 899 |
+
"grad_norm": 2.030813694000244,
|
| 900 |
+
"learning_rate": 2e-05,
|
| 901 |
+
"loss": 3.6907,
|
| 902 |
+
"step": 172500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.4602218851376103,
|
| 906 |
+
"grad_norm": 1.4791704416275024,
|
| 907 |
+
"learning_rate": 2e-05,
|
| 908 |
+
"loss": 3.6929,
|
| 909 |
+
"step": 174000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.46418931518190004,
|
| 913 |
+
"grad_norm": 1.747860312461853,
|
| 914 |
+
"learning_rate": 2e-05,
|
| 915 |
+
"loss": 3.6912,
|
| 916 |
+
"step": 175500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.4681567452261898,
|
| 920 |
+
"grad_norm": 1.837263584136963,
|
| 921 |
+
"learning_rate": 2e-05,
|
| 922 |
+
"loss": 3.6975,
|
| 923 |
+
"step": 177000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.47212417527047956,
|
| 927 |
+
"grad_norm": 1.7791885137557983,
|
| 928 |
+
"learning_rate": 2e-05,
|
| 929 |
+
"loss": 3.687,
|
| 930 |
+
"step": 178500
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.4760916053147693,
|
| 934 |
+
"grad_norm": 1.7485198974609375,
|
| 935 |
+
"learning_rate": 2e-05,
|
| 936 |
+
"loss": 3.6875,
|
| 937 |
+
"step": 180000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.4760916053147693,
|
| 941 |
+
"eval_loss": 3.666386127471924,
|
| 942 |
+
"eval_runtime": 508.5011,
|
| 943 |
+
"eval_samples_per_second": 131.388,
|
| 944 |
+
"eval_steps_per_second": 32.848,
|
| 945 |
+
"step": 180000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.480059035359059,
|
| 949 |
+
"grad_norm": 1.6664392948150635,
|
| 950 |
+
"learning_rate": 2e-05,
|
| 951 |
+
"loss": 3.6763,
|
| 952 |
+
"step": 181500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.48402646540334876,
|
| 956 |
+
"grad_norm": 1.7725111246109009,
|
| 957 |
+
"learning_rate": 2e-05,
|
| 958 |
+
"loss": 3.6824,
|
| 959 |
+
"step": 183000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.4879938954476385,
|
| 963 |
+
"grad_norm": 2.1763925552368164,
|
| 964 |
+
"learning_rate": 2e-05,
|
| 965 |
+
"loss": 3.6885,
|
| 966 |
+
"step": 184500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.4919613254919283,
|
| 970 |
+
"grad_norm": 1.492720603942871,
|
| 971 |
+
"learning_rate": 2e-05,
|
| 972 |
+
"loss": 3.6825,
|
| 973 |
+
"step": 186000
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.495928755536218,
|
| 977 |
+
"grad_norm": 1.365192174911499,
|
| 978 |
+
"learning_rate": 2e-05,
|
| 979 |
+
"loss": 3.6879,
|
| 980 |
+
"step": 187500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.49989618558050775,
|
| 984 |
+
"grad_norm": 1.8424572944641113,
|
| 985 |
+
"learning_rate": 2e-05,
|
| 986 |
+
"loss": 3.6825,
|
| 987 |
+
"step": 189000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 0.5038636156247975,
|
| 991 |
+
"grad_norm": 1.6156889200210571,
|
| 992 |
+
"learning_rate": 2e-05,
|
| 993 |
+
"loss": 3.6819,
|
| 994 |
+
"step": 190500
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.5078310456690872,
|
| 998 |
+
"grad_norm": 1.3726621866226196,
|
| 999 |
+
"learning_rate": 2e-05,
|
| 1000 |
+
"loss": 3.6766,
|
| 1001 |
+
"step": 192000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.511798475713377,
|
| 1005 |
+
"grad_norm": 1.5357227325439453,
|
| 1006 |
+
"learning_rate": 2e-05,
|
| 1007 |
+
"loss": 3.6686,
|
| 1008 |
+
"step": 193500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.5157659057576667,
|
| 1012 |
+
"grad_norm": 1.6667983531951904,
|
| 1013 |
+
"learning_rate": 2e-05,
|
| 1014 |
+
"loss": 3.673,
|
| 1015 |
+
"step": 195000
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.5157659057576667,
|
| 1019 |
+
"eval_loss": 3.6544015407562256,
|
| 1020 |
+
"eval_runtime": 508.3154,
|
| 1021 |
+
"eval_samples_per_second": 131.436,
|
| 1022 |
+
"eval_steps_per_second": 32.86,
|
| 1023 |
+
"step": 195000
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.5197333358019565,
|
| 1027 |
+
"grad_norm": 1.6955657005310059,
|
| 1028 |
+
"learning_rate": 2e-05,
|
| 1029 |
+
"loss": 3.6681,
|
| 1030 |
+
"step": 196500
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.5237007658462463,
|
| 1034 |
+
"grad_norm": 2.2115893363952637,
|
| 1035 |
+
"learning_rate": 2e-05,
|
| 1036 |
+
"loss": 3.6631,
|
| 1037 |
+
"step": 198000
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.5276681958905359,
|
| 1041 |
+
"grad_norm": 1.6372627019882202,
|
| 1042 |
+
"learning_rate": 2e-05,
|
| 1043 |
+
"loss": 3.6663,
|
| 1044 |
+
"step": 199500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.5316356259348257,
|
| 1048 |
+
"grad_norm": 1.719591498374939,
|
| 1049 |
+
"learning_rate": 2e-05,
|
| 1050 |
+
"loss": 3.6668,
|
| 1051 |
+
"step": 201000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.5356030559791154,
|
| 1055 |
+
"grad_norm": 1.9526976346969604,
|
| 1056 |
+
"learning_rate": 2e-05,
|
| 1057 |
+
"loss": 3.6614,
|
| 1058 |
+
"step": 202500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.5395704860234052,
|
| 1062 |
+
"grad_norm": 1.4962718486785889,
|
| 1063 |
+
"learning_rate": 2e-05,
|
| 1064 |
+
"loss": 3.6541,
|
| 1065 |
+
"step": 204000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.543537916067695,
|
| 1069 |
+
"grad_norm": 1.552954077720642,
|
| 1070 |
+
"learning_rate": 2e-05,
|
| 1071 |
+
"loss": 3.667,
|
| 1072 |
+
"step": 205500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.5475053461119846,
|
| 1076 |
+
"grad_norm": 2.0610477924346924,
|
| 1077 |
+
"learning_rate": 2e-05,
|
| 1078 |
+
"loss": 3.6551,
|
| 1079 |
+
"step": 207000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.5514727761562744,
|
| 1083 |
+
"grad_norm": 1.989310383796692,
|
| 1084 |
+
"learning_rate": 2e-05,
|
| 1085 |
+
"loss": 3.6638,
|
| 1086 |
+
"step": 208500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.5554402062005642,
|
| 1090 |
+
"grad_norm": 1.2716268301010132,
|
| 1091 |
+
"learning_rate": 2e-05,
|
| 1092 |
+
"loss": 3.6599,
|
| 1093 |
+
"step": 210000
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.5554402062005642,
|
| 1097 |
+
"eval_loss": 3.6426613330841064,
|
| 1098 |
+
"eval_runtime": 508.1725,
|
| 1099 |
+
"eval_samples_per_second": 131.473,
|
| 1100 |
+
"eval_steps_per_second": 32.869,
|
| 1101 |
+
"step": 210000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5594076362448539,
|
| 1105 |
+
"grad_norm": 1.81135892868042,
|
| 1106 |
+
"learning_rate": 2e-05,
|
| 1107 |
+
"loss": 3.6569,
|
| 1108 |
+
"step": 211500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5633750662891437,
|
| 1112 |
+
"grad_norm": 1.3680062294006348,
|
| 1113 |
+
"learning_rate": 2e-05,
|
| 1114 |
+
"loss": 3.6584,
|
| 1115 |
+
"step": 213000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5673424963334334,
|
| 1119 |
+
"grad_norm": 1.6358596086502075,
|
| 1120 |
+
"learning_rate": 2e-05,
|
| 1121 |
+
"loss": 3.6474,
|
| 1122 |
+
"step": 214500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5713099263777232,
|
| 1126 |
+
"grad_norm": 1.3548320531845093,
|
| 1127 |
+
"learning_rate": 2e-05,
|
| 1128 |
+
"loss": 3.6492,
|
| 1129 |
+
"step": 216000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5752773564220129,
|
| 1133 |
+
"grad_norm": 1.5863239765167236,
|
| 1134 |
+
"learning_rate": 2e-05,
|
| 1135 |
+
"loss": 3.6549,
|
| 1136 |
+
"step": 217500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5792447864663026,
|
| 1140 |
+
"grad_norm": 1.269634485244751,
|
| 1141 |
+
"learning_rate": 2e-05,
|
| 1142 |
+
"loss": 3.6405,
|
| 1143 |
+
"step": 219000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5832122165105924,
|
| 1147 |
+
"grad_norm": 1.405115008354187,
|
| 1148 |
+
"learning_rate": 2e-05,
|
| 1149 |
+
"loss": 3.6517,
|
| 1150 |
+
"step": 220500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5871796465548821,
|
| 1154 |
+
"grad_norm": 1.9946074485778809,
|
| 1155 |
+
"learning_rate": 2e-05,
|
| 1156 |
+
"loss": 3.6555,
|
| 1157 |
+
"step": 222000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5911470765991719,
|
| 1161 |
+
"grad_norm": 1.3142012357711792,
|
| 1162 |
+
"learning_rate": 2e-05,
|
| 1163 |
+
"loss": 3.6596,
|
| 1164 |
+
"step": 223500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5951145066434617,
|
| 1168 |
+
"grad_norm": 2.1748433113098145,
|
| 1169 |
+
"learning_rate": 2e-05,
|
| 1170 |
+
"loss": 3.6562,
|
| 1171 |
+
"step": 225000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5951145066434617,
|
| 1175 |
+
"eval_loss": 3.6325650215148926,
|
| 1176 |
+
"eval_runtime": 508.3225,
|
| 1177 |
+
"eval_samples_per_second": 131.434,
|
| 1178 |
+
"eval_steps_per_second": 32.859,
|
| 1179 |
+
"step": 225000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5990819366877513,
|
| 1183 |
+
"grad_norm": 1.736243724822998,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 3.6535,
|
| 1186 |
+
"step": 226500
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6030493667320411,
|
| 1190 |
+
"grad_norm": 1.4503512382507324,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 3.6472,
|
| 1193 |
+
"step": 228000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6070167967763308,
|
| 1197 |
+
"grad_norm": 1.7718149423599243,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 3.652,
|
| 1200 |
+
"step": 229500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6109842268206206,
|
| 1204 |
+
"grad_norm": 1.7987645864486694,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 3.6393,
|
| 1207 |
+
"step": 231000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.6149516568649104,
|
| 1211 |
+
"grad_norm": 2.3214948177337646,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 3.6552,
|
| 1214 |
+
"step": 232500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6189190869092001,
|
| 1218 |
+
"grad_norm": 1.3199760913848877,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 3.6537,
|
| 1221 |
+
"step": 234000
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6228865169534898,
|
| 1225 |
+
"grad_norm": 2.081068992614746,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 3.6401,
|
| 1228 |
+
"step": 235500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6268539469977795,
|
| 1232 |
+
"grad_norm": 1.275061845779419,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 3.6443,
|
| 1235 |
+
"step": 237000
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6308213770420693,
|
| 1239 |
+
"grad_norm": 1.7260355949401855,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 3.6342,
|
| 1242 |
+
"step": 238500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6347888070863591,
|
| 1246 |
+
"grad_norm": 1.4915698766708374,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 3.6409,
|
| 1249 |
+
"step": 240000
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6347888070863591,
|
| 1253 |
+
"eval_loss": 3.6224632263183594,
|
| 1254 |
+
"eval_runtime": 509.0689,
|
| 1255 |
+
"eval_samples_per_second": 131.242,
|
| 1256 |
+
"eval_steps_per_second": 32.811,
|
| 1257 |
+
"step": 240000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.6387562371306488,
|
| 1261 |
+
"grad_norm": 2.0755701065063477,
|
| 1262 |
+
"learning_rate": 2e-05,
|
| 1263 |
+
"loss": 3.6467,
|
| 1264 |
+
"step": 241500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.6427236671749386,
|
| 1268 |
+
"grad_norm": 1.1929559707641602,
|
| 1269 |
+
"learning_rate": 2e-05,
|
| 1270 |
+
"loss": 3.6437,
|
| 1271 |
+
"step": 243000
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.6466910972192282,
|
| 1275 |
+
"grad_norm": 1.8279653787612915,
|
| 1276 |
+
"learning_rate": 2e-05,
|
| 1277 |
+
"loss": 3.6362,
|
| 1278 |
+
"step": 244500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.650658527263518,
|
| 1282 |
+
"grad_norm": 2.0271174907684326,
|
| 1283 |
+
"learning_rate": 2e-05,
|
| 1284 |
+
"loss": 3.6278,
|
| 1285 |
+
"step": 246000
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.6546259573078078,
|
| 1289 |
+
"grad_norm": 2.5957841873168945,
|
| 1290 |
+
"learning_rate": 2e-05,
|
| 1291 |
+
"loss": 3.6292,
|
| 1292 |
+
"step": 247500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.6585933873520975,
|
| 1296 |
+
"grad_norm": 1.8614776134490967,
|
| 1297 |
+
"learning_rate": 2e-05,
|
| 1298 |
+
"loss": 3.6334,
|
| 1299 |
+
"step": 249000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.6625608173963873,
|
| 1303 |
+
"grad_norm": 1.9923863410949707,
|
| 1304 |
+
"learning_rate": 2e-05,
|
| 1305 |
+
"loss": 3.6389,
|
| 1306 |
+
"step": 250500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.666528247440677,
|
| 1310 |
+
"grad_norm": 1.719926357269287,
|
| 1311 |
+
"learning_rate": 2e-05,
|
| 1312 |
+
"loss": 3.6319,
|
| 1313 |
+
"step": 252000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.6704956774849667,
|
| 1317 |
+
"grad_norm": 1.542592167854309,
|
| 1318 |
+
"learning_rate": 2e-05,
|
| 1319 |
+
"loss": 3.6308,
|
| 1320 |
+
"step": 253500
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.6744631075292565,
|
| 1324 |
+
"grad_norm": 1.6688200235366821,
|
| 1325 |
+
"learning_rate": 2e-05,
|
| 1326 |
+
"loss": 3.6306,
|
| 1327 |
+
"step": 255000
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.6744631075292565,
|
| 1331 |
+
"eval_loss": 3.6125638484954834,
|
| 1332 |
+
"eval_runtime": 508.6624,
|
| 1333 |
+
"eval_samples_per_second": 131.346,
|
| 1334 |
+
"eval_steps_per_second": 32.837,
|
| 1335 |
+
"step": 255000
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 0.6784305375735462,
|
| 1339 |
+
"grad_norm": 2.0490431785583496,
|
| 1340 |
+
"learning_rate": 2e-05,
|
| 1341 |
+
"loss": 3.6257,
|
| 1342 |
+
"step": 256500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 0.682397967617836,
|
| 1346 |
+
"grad_norm": 1.5656068325042725,
|
| 1347 |
+
"learning_rate": 2e-05,
|
| 1348 |
+
"loss": 3.6327,
|
| 1349 |
+
"step": 258000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 0.6863653976621257,
|
| 1353 |
+
"grad_norm": 1.741310715675354,
|
| 1354 |
+
"learning_rate": 2e-05,
|
| 1355 |
+
"loss": 3.624,
|
| 1356 |
+
"step": 259500
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 0.6903328277064155,
|
| 1360 |
+
"grad_norm": 1.5890847444534302,
|
| 1361 |
+
"learning_rate": 2e-05,
|
| 1362 |
+
"loss": 3.631,
|
| 1363 |
+
"step": 261000
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 0.6943002577507053,
|
| 1367 |
+
"grad_norm": 2.6518447399139404,
|
| 1368 |
+
"learning_rate": 2e-05,
|
| 1369 |
+
"loss": 3.6219,
|
| 1370 |
+
"step": 262500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.6982676877949949,
|
| 1374 |
+
"grad_norm": 1.344978928565979,
|
| 1375 |
+
"learning_rate": 2e-05,
|
| 1376 |
+
"loss": 3.637,
|
| 1377 |
+
"step": 264000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 0.7022351178392847,
|
| 1381 |
+
"grad_norm": 1.8812129497528076,
|
| 1382 |
+
"learning_rate": 2e-05,
|
| 1383 |
+
"loss": 3.6273,
|
| 1384 |
+
"step": 265500
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 0.7062025478835744,
|
| 1388 |
+
"grad_norm": 1.59878408908844,
|
| 1389 |
+
"learning_rate": 2e-05,
|
| 1390 |
+
"loss": 3.6204,
|
| 1391 |
+
"step": 267000
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 0.7101699779278642,
|
| 1395 |
+
"grad_norm": 2.1738855838775635,
|
| 1396 |
+
"learning_rate": 2e-05,
|
| 1397 |
+
"loss": 3.6113,
|
| 1398 |
+
"step": 268500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.714137407972154,
|
| 1402 |
+
"grad_norm": 1.292671799659729,
|
| 1403 |
+
"learning_rate": 2e-05,
|
| 1404 |
+
"loss": 3.6244,
|
| 1405 |
+
"step": 270000
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 0.714137407972154,
|
| 1409 |
+
"eval_loss": 3.6037514209747314,
|
| 1410 |
+
"eval_runtime": 508.7856,
|
| 1411 |
+
"eval_samples_per_second": 131.315,
|
| 1412 |
+
"eval_steps_per_second": 32.829,
|
| 1413 |
+
"step": 270000
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 0.7181048380164436,
|
| 1417 |
+
"grad_norm": 1.5464812517166138,
|
| 1418 |
+
"learning_rate": 2e-05,
|
| 1419 |
+
"loss": 3.618,
|
| 1420 |
+
"step": 271500
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.7220722680607334,
|
| 1424 |
+
"grad_norm": 1.2013261318206787,
|
| 1425 |
+
"learning_rate": 2e-05,
|
| 1426 |
+
"loss": 3.6131,
|
| 1427 |
+
"step": 273000
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.7260396981050231,
|
| 1431 |
+
"grad_norm": 1.424142599105835,
|
| 1432 |
+
"learning_rate": 2e-05,
|
| 1433 |
+
"loss": 3.6223,
|
| 1434 |
+
"step": 274500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.7300071281493129,
|
| 1438 |
+
"grad_norm": 2.2044925689697266,
|
| 1439 |
+
"learning_rate": 2e-05,
|
| 1440 |
+
"loss": 3.6141,
|
| 1441 |
+
"step": 276000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.7339745581936027,
|
| 1445 |
+
"grad_norm": 1.668161392211914,
|
| 1446 |
+
"learning_rate": 2e-05,
|
| 1447 |
+
"loss": 3.6179,
|
| 1448 |
+
"step": 277500
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.7379419882378924,
|
| 1452 |
+
"grad_norm": 2.1473095417022705,
|
| 1453 |
+
"learning_rate": 2e-05,
|
| 1454 |
+
"loss": 3.6185,
|
| 1455 |
+
"step": 279000
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 0.7419094182821822,
|
| 1459 |
+
"grad_norm": 2.150111436843872,
|
| 1460 |
+
"learning_rate": 2e-05,
|
| 1461 |
+
"loss": 3.6244,
|
| 1462 |
+
"step": 280500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 0.7458768483264718,
|
| 1466 |
+
"grad_norm": 1.4405452013015747,
|
| 1467 |
+
"learning_rate": 2e-05,
|
| 1468 |
+
"loss": 3.6174,
|
| 1469 |
+
"step": 282000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 0.7498442783707616,
|
| 1473 |
+
"grad_norm": 1.622565507888794,
|
| 1474 |
+
"learning_rate": 2e-05,
|
| 1475 |
+
"loss": 3.6176,
|
| 1476 |
+
"step": 283500
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 0.7538117084150514,
|
| 1480 |
+
"grad_norm": 1.6604636907577515,
|
| 1481 |
+
"learning_rate": 2e-05,
|
| 1482 |
+
"loss": 3.6026,
|
| 1483 |
+
"step": 285000
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 0.7538117084150514,
|
| 1487 |
+
"eval_loss": 3.594251871109009,
|
| 1488 |
+
"eval_runtime": 508.3923,
|
| 1489 |
+
"eval_samples_per_second": 131.416,
|
| 1490 |
+
"eval_steps_per_second": 32.855,
|
| 1491 |
+
"step": 285000
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 0.7577791384593411,
|
| 1495 |
+
"grad_norm": 1.5415109395980835,
|
| 1496 |
+
"learning_rate": 2e-05,
|
| 1497 |
+
"loss": 3.6221,
|
| 1498 |
+
"step": 286500
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.7617465685036309,
|
| 1502 |
+
"grad_norm": 2.0804574489593506,
|
| 1503 |
+
"learning_rate": 2e-05,
|
| 1504 |
+
"loss": 3.6095,
|
| 1505 |
+
"step": 288000
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.7657139985479205,
|
| 1509 |
+
"grad_norm": 1.6679505109786987,
|
| 1510 |
+
"learning_rate": 2e-05,
|
| 1511 |
+
"loss": 3.6175,
|
| 1512 |
+
"step": 289500
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 0.7696814285922103,
|
| 1516 |
+
"grad_norm": 1.562348484992981,
|
| 1517 |
+
"learning_rate": 2e-05,
|
| 1518 |
+
"loss": 3.604,
|
| 1519 |
+
"step": 291000
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 0.7736488586365001,
|
| 1523 |
+
"grad_norm": 1.4263495206832886,
|
| 1524 |
+
"learning_rate": 2e-05,
|
| 1525 |
+
"loss": 3.6253,
|
| 1526 |
+
"step": 292500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.7776162886807898,
|
| 1530 |
+
"grad_norm": 2.324164390563965,
|
| 1531 |
+
"learning_rate": 2e-05,
|
| 1532 |
+
"loss": 3.6093,
|
| 1533 |
+
"step": 294000
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 0.7815837187250796,
|
| 1537 |
+
"grad_norm": 1.5391567945480347,
|
| 1538 |
+
"learning_rate": 2e-05,
|
| 1539 |
+
"loss": 3.5999,
|
| 1540 |
+
"step": 295500
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.7855511487693694,
|
| 1544 |
+
"grad_norm": 1.6802319288253784,
|
| 1545 |
+
"learning_rate": 2e-05,
|
| 1546 |
+
"loss": 3.6148,
|
| 1547 |
+
"step": 297000
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 0.789518578813659,
|
| 1551 |
+
"grad_norm": 1.5229837894439697,
|
| 1552 |
+
"learning_rate": 2e-05,
|
| 1553 |
+
"loss": 3.6033,
|
| 1554 |
+
"step": 298500
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.7934860088579488,
|
| 1558 |
+
"grad_norm": 1.6937259435653687,
|
| 1559 |
+
"learning_rate": 2e-05,
|
| 1560 |
+
"loss": 3.606,
|
| 1561 |
+
"step": 300000
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.7934860088579488,
|
| 1565 |
+
"eval_loss": 3.5862793922424316,
|
| 1566 |
+
"eval_runtime": 512.2421,
|
| 1567 |
+
"eval_samples_per_second": 130.429,
|
| 1568 |
+
"eval_steps_per_second": 32.608,
|
| 1569 |
+
"step": 300000
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.7974534389022385,
|
| 1573 |
+
"grad_norm": 1.4837462902069092,
|
| 1574 |
+
"learning_rate": 2e-05,
|
| 1575 |
+
"loss": 3.6056,
|
| 1576 |
+
"step": 301500
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.8014208689465283,
|
| 1580 |
+
"grad_norm": 2.499969482421875,
|
| 1581 |
+
"learning_rate": 2e-05,
|
| 1582 |
+
"loss": 3.5886,
|
| 1583 |
+
"step": 303000
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 0.8053882989908181,
|
| 1587 |
+
"grad_norm": 1.9505125284194946,
|
| 1588 |
+
"learning_rate": 2e-05,
|
| 1589 |
+
"loss": 3.6012,
|
| 1590 |
+
"step": 304500
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.8093557290351078,
|
| 1594 |
+
"grad_norm": 1.3550680875778198,
|
| 1595 |
+
"learning_rate": 2e-05,
|
| 1596 |
+
"loss": 3.604,
|
| 1597 |
+
"step": 306000
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 0.8133231590793976,
|
| 1601 |
+
"grad_norm": 1.6309428215026855,
|
| 1602 |
+
"learning_rate": 2e-05,
|
| 1603 |
+
"loss": 3.5924,
|
| 1604 |
+
"step": 307500
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.8172905891236872,
|
| 1608 |
+
"grad_norm": 1.4535800218582153,
|
| 1609 |
+
"learning_rate": 2e-05,
|
| 1610 |
+
"loss": 3.6027,
|
| 1611 |
+
"step": 309000
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 0.821258019167977,
|
| 1615 |
+
"grad_norm": 1.4342435598373413,
|
| 1616 |
+
"learning_rate": 2e-05,
|
| 1617 |
+
"loss": 3.5984,
|
| 1618 |
+
"step": 310500
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 0.8252254492122668,
|
| 1622 |
+
"grad_norm": 1.4876697063446045,
|
| 1623 |
+
"learning_rate": 2e-05,
|
| 1624 |
+
"loss": 3.608,
|
| 1625 |
+
"step": 312000
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.8291928792565565,
|
| 1629 |
+
"grad_norm": 1.625393271446228,
|
| 1630 |
+
"learning_rate": 2e-05,
|
| 1631 |
+
"loss": 3.5937,
|
| 1632 |
+
"step": 313500
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.8331603093008463,
|
| 1636 |
+
"grad_norm": 1.6351014375686646,
|
| 1637 |
+
"learning_rate": 2e-05,
|
| 1638 |
+
"loss": 3.5954,
|
| 1639 |
+
"step": 315000
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 0.8331603093008463,
|
| 1643 |
+
"eval_loss": 3.578728675842285,
|
| 1644 |
+
"eval_runtime": 509.839,
|
| 1645 |
+
"eval_samples_per_second": 131.043,
|
| 1646 |
+
"eval_steps_per_second": 32.761,
|
| 1647 |
+
"step": 315000
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.837127739345136,
|
| 1651 |
+
"grad_norm": 1.5755149126052856,
|
| 1652 |
+
"learning_rate": 2e-05,
|
| 1653 |
+
"loss": 3.5986,
|
| 1654 |
+
"step": 316500
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8410951693894257,
|
| 1658 |
+
"grad_norm": 2.0601248741149902,
|
| 1659 |
+
"learning_rate": 2e-05,
|
| 1660 |
+
"loss": 3.5942,
|
| 1661 |
+
"step": 318000
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8450625994337155,
|
| 1665 |
+
"grad_norm": 1.9317686557769775,
|
| 1666 |
+
"learning_rate": 2e-05,
|
| 1667 |
+
"loss": 3.5993,
|
| 1668 |
+
"step": 319500
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8490300294780052,
|
| 1672 |
+
"grad_norm": 2.375718593597412,
|
| 1673 |
+
"learning_rate": 2e-05,
|
| 1674 |
+
"loss": 3.5833,
|
| 1675 |
+
"step": 321000
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.852997459522295,
|
| 1679 |
+
"grad_norm": 2.2386367321014404,
|
| 1680 |
+
"learning_rate": 2e-05,
|
| 1681 |
+
"loss": 3.587,
|
| 1682 |
+
"step": 322500
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8569648895665847,
|
| 1686 |
+
"grad_norm": 1.9626859426498413,
|
| 1687 |
+
"learning_rate": 2e-05,
|
| 1688 |
+
"loss": 3.5951,
|
| 1689 |
+
"step": 324000
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8609323196108745,
|
| 1693 |
+
"grad_norm": 1.279280424118042,
|
| 1694 |
+
"learning_rate": 2e-05,
|
| 1695 |
+
"loss": 3.5881,
|
| 1696 |
+
"step": 325500
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8648997496551643,
|
| 1700 |
+
"grad_norm": 1.8277925252914429,
|
| 1701 |
+
"learning_rate": 2e-05,
|
| 1702 |
+
"loss": 3.5933,
|
| 1703 |
+
"step": 327000
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8688671796994539,
|
| 1707 |
+
"grad_norm": 1.578514575958252,
|
| 1708 |
+
"learning_rate": 2e-05,
|
| 1709 |
+
"loss": 3.5936,
|
| 1710 |
+
"step": 328500
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8728346097437437,
|
| 1714 |
+
"grad_norm": 1.8757489919662476,
|
| 1715 |
+
"learning_rate": 2e-05,
|
| 1716 |
+
"loss": 3.5997,
|
| 1717 |
+
"step": 330000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8728346097437437,
|
| 1721 |
+
"eval_loss": 3.5698554515838623,
|
| 1722 |
+
"eval_runtime": 508.3917,
|
| 1723 |
+
"eval_samples_per_second": 131.416,
|
| 1724 |
+
"eval_steps_per_second": 32.855,
|
| 1725 |
+
"step": 330000
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.8768020397880334,
|
| 1729 |
+
"grad_norm": 1.7685048580169678,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 3.5876,
|
| 1732 |
+
"step": 331500
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.8807694698323232,
|
| 1736 |
+
"grad_norm": 1.6386830806732178,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 3.5908,
|
| 1739 |
+
"step": 333000
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.884736899876613,
|
| 1743 |
+
"grad_norm": 1.6260840892791748,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 3.5839,
|
| 1746 |
+
"step": 334500
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.8887043299209026,
|
| 1750 |
+
"grad_norm": 1.721689224243164,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 3.5876,
|
| 1753 |
+
"step": 336000
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.8926717599651924,
|
| 1757 |
+
"grad_norm": 2.114492654800415,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 3.5752,
|
| 1760 |
+
"step": 337500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.8966391900094821,
|
| 1764 |
+
"grad_norm": 1.563759446144104,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 3.5679,
|
| 1767 |
+
"step": 339000
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9006066200537719,
|
| 1771 |
+
"grad_norm": 1.5705732107162476,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 3.5934,
|
| 1774 |
+
"step": 340500
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.9045740500980617,
|
| 1778 |
+
"grad_norm": 1.3177400827407837,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 3.5942,
|
| 1781 |
+
"step": 342000
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9085414801423514,
|
| 1785 |
+
"grad_norm": 2.3175089359283447,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 3.5812,
|
| 1788 |
+
"step": 343500
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9125089101866412,
|
| 1792 |
+
"grad_norm": 1.7788249254226685,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 3.5801,
|
| 1795 |
+
"step": 345000
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9125089101866412,
|
| 1799 |
+
"eval_loss": 3.562117338180542,
|
| 1800 |
+
"eval_runtime": 508.9343,
|
| 1801 |
+
"eval_samples_per_second": 131.276,
|
| 1802 |
+
"eval_steps_per_second": 32.82,
|
| 1803 |
+
"step": 345000
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.9164763402309308,
|
| 1807 |
+
"grad_norm": 1.8623145818710327,
|
| 1808 |
+
"learning_rate": 2e-05,
|
| 1809 |
+
"loss": 3.5826,
|
| 1810 |
+
"step": 346500
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.9204437702752206,
|
| 1814 |
+
"grad_norm": 1.7332490682601929,
|
| 1815 |
+
"learning_rate": 2e-05,
|
| 1816 |
+
"loss": 3.5857,
|
| 1817 |
+
"step": 348000
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.9244112003195104,
|
| 1821 |
+
"grad_norm": 2.2232964038848877,
|
| 1822 |
+
"learning_rate": 2e-05,
|
| 1823 |
+
"loss": 3.5789,
|
| 1824 |
+
"step": 349500
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.9283786303638001,
|
| 1828 |
+
"grad_norm": 1.4348344802856445,
|
| 1829 |
+
"learning_rate": 2e-05,
|
| 1830 |
+
"loss": 3.5736,
|
| 1831 |
+
"step": 351000
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.9323460604080899,
|
| 1835 |
+
"grad_norm": 1.3817743062973022,
|
| 1836 |
+
"learning_rate": 2e-05,
|
| 1837 |
+
"loss": 3.5759,
|
| 1838 |
+
"step": 352500
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.9363134904523795,
|
| 1842 |
+
"grad_norm": 1.8987672328948975,
|
| 1843 |
+
"learning_rate": 2e-05,
|
| 1844 |
+
"loss": 3.5743,
|
| 1845 |
+
"step": 354000
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.9402809204966693,
|
| 1849 |
+
"grad_norm": 1.5825597047805786,
|
| 1850 |
+
"learning_rate": 2e-05,
|
| 1851 |
+
"loss": 3.5746,
|
| 1852 |
+
"step": 355500
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.9442483505409591,
|
| 1856 |
+
"grad_norm": 1.5452762842178345,
|
| 1857 |
+
"learning_rate": 2e-05,
|
| 1858 |
+
"loss": 3.5781,
|
| 1859 |
+
"step": 357000
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.9482157805852488,
|
| 1863 |
+
"grad_norm": 1.3254438638687134,
|
| 1864 |
+
"learning_rate": 2e-05,
|
| 1865 |
+
"loss": 3.5868,
|
| 1866 |
+
"step": 358500
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.9521832106295386,
|
| 1870 |
+
"grad_norm": 2.779953956604004,
|
| 1871 |
+
"learning_rate": 2e-05,
|
| 1872 |
+
"loss": 3.5705,
|
| 1873 |
+
"step": 360000
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.9521832106295386,
|
| 1877 |
+
"eval_loss": 3.5557963848114014,
|
| 1878 |
+
"eval_runtime": 508.3414,
|
| 1879 |
+
"eval_samples_per_second": 131.429,
|
| 1880 |
+
"eval_steps_per_second": 32.858,
|
| 1881 |
+
"step": 360000
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 0.9561506406738283,
|
| 1885 |
+
"grad_norm": 1.8877593278884888,
|
| 1886 |
+
"learning_rate": 2e-05,
|
| 1887 |
+
"loss": 3.561,
|
| 1888 |
+
"step": 361500
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 0.960118070718118,
|
| 1892 |
+
"grad_norm": 1.5967822074890137,
|
| 1893 |
+
"learning_rate": 2e-05,
|
| 1894 |
+
"loss": 3.5633,
|
| 1895 |
+
"step": 363000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 0.9640855007624078,
|
| 1899 |
+
"grad_norm": 2.1027090549468994,
|
| 1900 |
+
"learning_rate": 2e-05,
|
| 1901 |
+
"loss": 3.576,
|
| 1902 |
+
"step": 364500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 0.9680529308066975,
|
| 1906 |
+
"grad_norm": 1.8715656995773315,
|
| 1907 |
+
"learning_rate": 2e-05,
|
| 1908 |
+
"loss": 3.5809,
|
| 1909 |
+
"step": 366000
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 0.9720203608509873,
|
| 1913 |
+
"grad_norm": 1.6980156898498535,
|
| 1914 |
+
"learning_rate": 2e-05,
|
| 1915 |
+
"loss": 3.5678,
|
| 1916 |
+
"step": 367500
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 0.975987790895277,
|
| 1920 |
+
"grad_norm": 1.4774205684661865,
|
| 1921 |
+
"learning_rate": 2e-05,
|
| 1922 |
+
"loss": 3.5771,
|
| 1923 |
+
"step": 369000
|
| 1924 |
+
}
|
| 1925 |
+
],
|
| 1926 |
+
"logging_steps": 1500,
|
| 1927 |
+
"max_steps": 378078,
|
| 1928 |
+
"num_input_tokens_seen": 0,
|
| 1929 |
+
"num_train_epochs": 1,
|
| 1930 |
+
"save_steps": 10000,
|
| 1931 |
+
"stateful_callbacks": {
|
| 1932 |
+
"TrainerControl": {
|
| 1933 |
+
"args": {
|
| 1934 |
+
"should_epoch_stop": false,
|
| 1935 |
+
"should_evaluate": false,
|
| 1936 |
+
"should_log": false,
|
| 1937 |
+
"should_save": true,
|
| 1938 |
+
"should_training_stop": false
|
| 1939 |
+
},
|
| 1940 |
+
"attributes": {}
|
| 1941 |
+
}
|
| 1942 |
+
},
|
| 1943 |
+
"total_flos": 4.653629565827236e+18,
|
| 1944 |
+
"train_batch_size": 8,
|
| 1945 |
+
"trial_name": null,
|
| 1946 |
+
"trial_params": null
|
| 1947 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0002e7e084213a534b889c72e039fd6fc9abae9b0498d394c5c7323149eb7ad
|
| 3 |
+
size 5112
|