Upload folder using huggingface_hub
Browse files- adapter_config.json +34 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +1826 -0
- training_args.bin +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "nicholasKluge/TeenyTinyLlama-460m",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 16,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 8,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": [
|
| 27 |
+
"q_proj",
|
| 28 |
+
"v_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM",
|
| 31 |
+
"trainable_token_indices": null,
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e25edca5a6d7ddef8b3e30502982709a0b20a24cbd2d50169477550e925c01ca
|
| 3 |
+
size 3158328
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a97db60444fb96bd6f4e204c699a5a67ea0d6d984030af311b67aaeb003fb61
|
| 3 |
+
size 6372346
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77423b0becbc907159fd91a6c473c0ef4c0ce9b58c7ca7e0da090177da458454
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7eb9a23067606b7785cda6658070f2efc2013275bc7441e316922d1acbd4fdfe
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 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 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"3": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
}
|
| 38 |
+
},
|
| 39 |
+
"bos_token": "<s>",
|
| 40 |
+
"bos_token_id": 1,
|
| 41 |
+
"clean_up_tokenization_spaces": false,
|
| 42 |
+
"eos_token": "</s>",
|
| 43 |
+
"eos_token_id": 2,
|
| 44 |
+
"extra_special_tokens": {},
|
| 45 |
+
"legacy": false,
|
| 46 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 47 |
+
"pad_token": "</s>",
|
| 48 |
+
"pad_token_id": 2,
|
| 49 |
+
"padding_side": "right",
|
| 50 |
+
"sp_model_kwargs": {},
|
| 51 |
+
"tokenizer_class": "LlamaTokenizerFast",
|
| 52 |
+
"unk_token": "<unk>",
|
| 53 |
+
"unk_token_id": 0,
|
| 54 |
+
"use_default_system_prompt": false
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1826 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.9975031210986267,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 12800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007802746566791511,
|
| 14 |
+
"grad_norm": 0.09780355542898178,
|
| 15 |
+
"learning_rate": 4.9923533083645446e-05,
|
| 16 |
+
"loss": 2.1517,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.015605493133583021,
|
| 21 |
+
"grad_norm": 0.1110800951719284,
|
| 22 |
+
"learning_rate": 4.984550561797753e-05,
|
| 23 |
+
"loss": 2.0733,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.023408239700374533,
|
| 28 |
+
"grad_norm": 0.11351309716701508,
|
| 29 |
+
"learning_rate": 4.9767478152309616e-05,
|
| 30 |
+
"loss": 1.9999,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.031210986267166042,
|
| 35 |
+
"grad_norm": 0.12184558063745499,
|
| 36 |
+
"learning_rate": 4.96894506866417e-05,
|
| 37 |
+
"loss": 1.911,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03901373283395755,
|
| 42 |
+
"grad_norm": 0.12001396715641022,
|
| 43 |
+
"learning_rate": 4.9611423220973786e-05,
|
| 44 |
+
"loss": 1.8957,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04681647940074907,
|
| 49 |
+
"grad_norm": 0.13359545171260834,
|
| 50 |
+
"learning_rate": 4.9533395755305875e-05,
|
| 51 |
+
"loss": 1.881,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.054619225967540576,
|
| 56 |
+
"grad_norm": 0.177729532122612,
|
| 57 |
+
"learning_rate": 4.9455368289637956e-05,
|
| 58 |
+
"loss": 1.8426,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.062421972534332085,
|
| 63 |
+
"grad_norm": 0.15538156032562256,
|
| 64 |
+
"learning_rate": 4.9377340823970044e-05,
|
| 65 |
+
"loss": 1.8528,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0702247191011236,
|
| 70 |
+
"grad_norm": 0.1490369588136673,
|
| 71 |
+
"learning_rate": 4.9299313358302126e-05,
|
| 72 |
+
"loss": 1.8604,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0780274656679151,
|
| 77 |
+
"grad_norm": 0.15821540355682373,
|
| 78 |
+
"learning_rate": 4.9221285892634214e-05,
|
| 79 |
+
"loss": 1.8454,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08583021223470662,
|
| 84 |
+
"grad_norm": 0.15961939096450806,
|
| 85 |
+
"learning_rate": 4.9143258426966296e-05,
|
| 86 |
+
"loss": 1.8235,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09363295880149813,
|
| 91 |
+
"grad_norm": 0.18454231321811676,
|
| 92 |
+
"learning_rate": 4.906523096129838e-05,
|
| 93 |
+
"loss": 1.8245,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10143570536828964,
|
| 98 |
+
"grad_norm": 0.19588346779346466,
|
| 99 |
+
"learning_rate": 4.8987203495630466e-05,
|
| 100 |
+
"loss": 1.8332,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10923845193508115,
|
| 105 |
+
"grad_norm": 0.20523308217525482,
|
| 106 |
+
"learning_rate": 4.890917602996255e-05,
|
| 107 |
+
"loss": 1.8176,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11704119850187265,
|
| 112 |
+
"grad_norm": 0.2048981934785843,
|
| 113 |
+
"learning_rate": 4.883114856429463e-05,
|
| 114 |
+
"loss": 1.7751,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12484394506866417,
|
| 119 |
+
"grad_norm": 0.19311609864234924,
|
| 120 |
+
"learning_rate": 4.875312109862672e-05,
|
| 121 |
+
"loss": 1.8034,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13264669163545567,
|
| 126 |
+
"grad_norm": 0.21123754978179932,
|
| 127 |
+
"learning_rate": 4.86750936329588e-05,
|
| 128 |
+
"loss": 1.7859,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1404494382022472,
|
| 133 |
+
"grad_norm": 0.270113080739975,
|
| 134 |
+
"learning_rate": 4.859706616729089e-05,
|
| 135 |
+
"loss": 1.8121,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1482521847690387,
|
| 140 |
+
"grad_norm": 0.23597899079322815,
|
| 141 |
+
"learning_rate": 4.8519038701622975e-05,
|
| 142 |
+
"loss": 1.8213,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1560549313358302,
|
| 147 |
+
"grad_norm": 0.2787221372127533,
|
| 148 |
+
"learning_rate": 4.844101123595506e-05,
|
| 149 |
+
"loss": 1.8048,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.16385767790262173,
|
| 154 |
+
"grad_norm": 0.2505972981452942,
|
| 155 |
+
"learning_rate": 4.8362983770287145e-05,
|
| 156 |
+
"loss": 1.7799,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.17166042446941324,
|
| 161 |
+
"grad_norm": 0.2511955797672272,
|
| 162 |
+
"learning_rate": 4.828495630461923e-05,
|
| 163 |
+
"loss": 1.7758,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.17946317103620474,
|
| 168 |
+
"grad_norm": 0.2740708291530609,
|
| 169 |
+
"learning_rate": 4.8206928838951315e-05,
|
| 170 |
+
"loss": 1.7817,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.18726591760299627,
|
| 175 |
+
"grad_norm": 0.30172184109687805,
|
| 176 |
+
"learning_rate": 4.81289013732834e-05,
|
| 177 |
+
"loss": 1.7665,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.19506866416978777,
|
| 182 |
+
"grad_norm": 0.2734954059123993,
|
| 183 |
+
"learning_rate": 4.8050873907615485e-05,
|
| 184 |
+
"loss": 1.7673,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.20287141073657927,
|
| 189 |
+
"grad_norm": 0.2837677597999573,
|
| 190 |
+
"learning_rate": 4.797284644194757e-05,
|
| 191 |
+
"loss": 1.7539,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.21067415730337077,
|
| 196 |
+
"grad_norm": 0.287616491317749,
|
| 197 |
+
"learning_rate": 4.7894818976279655e-05,
|
| 198 |
+
"loss": 1.7845,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2184769038701623,
|
| 203 |
+
"grad_norm": 0.35649779438972473,
|
| 204 |
+
"learning_rate": 4.7816791510611737e-05,
|
| 205 |
+
"loss": 1.7629,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2262796504369538,
|
| 210 |
+
"grad_norm": 0.3466143012046814,
|
| 211 |
+
"learning_rate": 4.7738764044943825e-05,
|
| 212 |
+
"loss": 1.7637,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2340823970037453,
|
| 217 |
+
"grad_norm": 0.29961591958999634,
|
| 218 |
+
"learning_rate": 4.7660736579275906e-05,
|
| 219 |
+
"loss": 1.7775,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.24188514357053684,
|
| 224 |
+
"grad_norm": 0.3261962831020355,
|
| 225 |
+
"learning_rate": 4.7582709113607995e-05,
|
| 226 |
+
"loss": 1.7651,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.24968789013732834,
|
| 231 |
+
"grad_norm": 0.31850898265838623,
|
| 232 |
+
"learning_rate": 4.7504681647940076e-05,
|
| 233 |
+
"loss": 1.7601,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.25749063670411987,
|
| 238 |
+
"grad_norm": 0.3334502577781677,
|
| 239 |
+
"learning_rate": 4.7426654182272165e-05,
|
| 240 |
+
"loss": 1.7783,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.26529338327091134,
|
| 245 |
+
"grad_norm": 0.3199051320552826,
|
| 246 |
+
"learning_rate": 4.7348626716604246e-05,
|
| 247 |
+
"loss": 1.7492,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2730961298377029,
|
| 252 |
+
"grad_norm": 0.3381502628326416,
|
| 253 |
+
"learning_rate": 4.7270599250936335e-05,
|
| 254 |
+
"loss": 1.7592,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2808988764044944,
|
| 259 |
+
"grad_norm": 0.33800897002220154,
|
| 260 |
+
"learning_rate": 4.7192571785268416e-05,
|
| 261 |
+
"loss": 1.7403,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2887016229712859,
|
| 266 |
+
"grad_norm": 0.3942790925502777,
|
| 267 |
+
"learning_rate": 4.71145443196005e-05,
|
| 268 |
+
"loss": 1.7955,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2965043695380774,
|
| 273 |
+
"grad_norm": 0.3514516353607178,
|
| 274 |
+
"learning_rate": 4.7036516853932586e-05,
|
| 275 |
+
"loss": 1.7453,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.30430711610486894,
|
| 280 |
+
"grad_norm": 0.3649444878101349,
|
| 281 |
+
"learning_rate": 4.695848938826467e-05,
|
| 282 |
+
"loss": 1.7584,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3121098626716604,
|
| 287 |
+
"grad_norm": 0.33589330315589905,
|
| 288 |
+
"learning_rate": 4.6880461922596756e-05,
|
| 289 |
+
"loss": 1.7369,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.31991260923845194,
|
| 294 |
+
"grad_norm": 0.3481082320213318,
|
| 295 |
+
"learning_rate": 4.680243445692884e-05,
|
| 296 |
+
"loss": 1.7281,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.32771535580524347,
|
| 301 |
+
"grad_norm": 0.34206053614616394,
|
| 302 |
+
"learning_rate": 4.6724406991260926e-05,
|
| 303 |
+
"loss": 1.7266,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.33551810237203494,
|
| 308 |
+
"grad_norm": 0.3519572615623474,
|
| 309 |
+
"learning_rate": 4.664637952559301e-05,
|
| 310 |
+
"loss": 1.7365,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3433208489388265,
|
| 315 |
+
"grad_norm": 0.35082659125328064,
|
| 316 |
+
"learning_rate": 4.6568352059925096e-05,
|
| 317 |
+
"loss": 1.7411,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.351123595505618,
|
| 322 |
+
"grad_norm": 0.3652777373790741,
|
| 323 |
+
"learning_rate": 4.649032459425718e-05,
|
| 324 |
+
"loss": 1.7466,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3589263420724095,
|
| 329 |
+
"grad_norm": 0.39173486828804016,
|
| 330 |
+
"learning_rate": 4.6412297128589266e-05,
|
| 331 |
+
"loss": 1.7443,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.366729088639201,
|
| 336 |
+
"grad_norm": 0.47386667132377625,
|
| 337 |
+
"learning_rate": 4.633426966292135e-05,
|
| 338 |
+
"loss": 1.7276,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.37453183520599254,
|
| 343 |
+
"grad_norm": 0.3906112015247345,
|
| 344 |
+
"learning_rate": 4.6256242197253436e-05,
|
| 345 |
+
"loss": 1.7194,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.382334581772784,
|
| 350 |
+
"grad_norm": 0.4129493832588196,
|
| 351 |
+
"learning_rate": 4.6178214731585524e-05,
|
| 352 |
+
"loss": 1.7203,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.39013732833957554,
|
| 357 |
+
"grad_norm": 0.4499260485172272,
|
| 358 |
+
"learning_rate": 4.6100187265917605e-05,
|
| 359 |
+
"loss": 1.7268,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.397940074906367,
|
| 364 |
+
"grad_norm": 0.4021989703178406,
|
| 365 |
+
"learning_rate": 4.6022159800249694e-05,
|
| 366 |
+
"loss": 1.7532,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.40574282147315854,
|
| 371 |
+
"grad_norm": 0.3842841386795044,
|
| 372 |
+
"learning_rate": 4.5944132334581775e-05,
|
| 373 |
+
"loss": 1.7195,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4135455680399501,
|
| 378 |
+
"grad_norm": 0.3462275266647339,
|
| 379 |
+
"learning_rate": 4.5866104868913864e-05,
|
| 380 |
+
"loss": 1.7269,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.42134831460674155,
|
| 385 |
+
"grad_norm": 0.39410606026649475,
|
| 386 |
+
"learning_rate": 4.5788077403245945e-05,
|
| 387 |
+
"loss": 1.7192,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4291510611735331,
|
| 392 |
+
"grad_norm": 0.4126410484313965,
|
| 393 |
+
"learning_rate": 4.5710049937578034e-05,
|
| 394 |
+
"loss": 1.7421,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4369538077403246,
|
| 399 |
+
"grad_norm": 0.36984720826148987,
|
| 400 |
+
"learning_rate": 4.5632022471910115e-05,
|
| 401 |
+
"loss": 1.7101,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4447565543071161,
|
| 406 |
+
"grad_norm": 0.3527930974960327,
|
| 407 |
+
"learning_rate": 4.5553995006242203e-05,
|
| 408 |
+
"loss": 1.7276,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4525593008739076,
|
| 413 |
+
"grad_norm": 0.4303431510925293,
|
| 414 |
+
"learning_rate": 4.5475967540574285e-05,
|
| 415 |
+
"loss": 1.7292,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.46036204744069914,
|
| 420 |
+
"grad_norm": 0.43609175086021423,
|
| 421 |
+
"learning_rate": 4.5397940074906367e-05,
|
| 422 |
+
"loss": 1.7167,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.4681647940074906,
|
| 427 |
+
"grad_norm": 0.43081963062286377,
|
| 428 |
+
"learning_rate": 4.5319912609238455e-05,
|
| 429 |
+
"loss": 1.7214,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.47596754057428214,
|
| 434 |
+
"grad_norm": 0.4468071162700653,
|
| 435 |
+
"learning_rate": 4.5241885143570536e-05,
|
| 436 |
+
"loss": 1.7375,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.4837702871410737,
|
| 441 |
+
"grad_norm": 0.4700297713279724,
|
| 442 |
+
"learning_rate": 4.516385767790262e-05,
|
| 443 |
+
"loss": 1.7245,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.49157303370786515,
|
| 448 |
+
"grad_norm": 0.4510380029678345,
|
| 449 |
+
"learning_rate": 4.5085830212234706e-05,
|
| 450 |
+
"loss": 1.7303,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.4993757802746567,
|
| 455 |
+
"grad_norm": 0.4726191461086273,
|
| 456 |
+
"learning_rate": 4.5007802746566795e-05,
|
| 457 |
+
"loss": 1.7146,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5071785268414482,
|
| 462 |
+
"grad_norm": 0.4493762254714966,
|
| 463 |
+
"learning_rate": 4.4929775280898876e-05,
|
| 464 |
+
"loss": 1.7214,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5149812734082397,
|
| 469 |
+
"grad_norm": 0.425413578748703,
|
| 470 |
+
"learning_rate": 4.4851747815230965e-05,
|
| 471 |
+
"loss": 1.6912,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5227840199750312,
|
| 476 |
+
"grad_norm": 0.49669891595840454,
|
| 477 |
+
"learning_rate": 4.4773720349563046e-05,
|
| 478 |
+
"loss": 1.7082,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5305867665418227,
|
| 483 |
+
"grad_norm": 0.5002058148384094,
|
| 484 |
+
"learning_rate": 4.4695692883895134e-05,
|
| 485 |
+
"loss": 1.7119,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5383895131086143,
|
| 490 |
+
"grad_norm": 0.5210412740707397,
|
| 491 |
+
"learning_rate": 4.4617665418227216e-05,
|
| 492 |
+
"loss": 1.7139,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5461922596754057,
|
| 497 |
+
"grad_norm": 0.42913952469825745,
|
| 498 |
+
"learning_rate": 4.4539637952559304e-05,
|
| 499 |
+
"loss": 1.7083,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5539950062421972,
|
| 504 |
+
"grad_norm": 0.45524224638938904,
|
| 505 |
+
"learning_rate": 4.4461610486891386e-05,
|
| 506 |
+
"loss": 1.715,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5617977528089888,
|
| 511 |
+
"grad_norm": 0.47845831513404846,
|
| 512 |
+
"learning_rate": 4.4383583021223474e-05,
|
| 513 |
+
"loss": 1.7194,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5696004993757803,
|
| 518 |
+
"grad_norm": 0.4836772680282593,
|
| 519 |
+
"learning_rate": 4.4305555555555556e-05,
|
| 520 |
+
"loss": 1.7009,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5774032459425718,
|
| 525 |
+
"grad_norm": 0.4728386700153351,
|
| 526 |
+
"learning_rate": 4.4227528089887644e-05,
|
| 527 |
+
"loss": 1.7128,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5852059925093633,
|
| 532 |
+
"grad_norm": 0.4861275255680084,
|
| 533 |
+
"learning_rate": 4.4149500624219726e-05,
|
| 534 |
+
"loss": 1.7089,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5930087390761548,
|
| 539 |
+
"grad_norm": 0.5425216555595398,
|
| 540 |
+
"learning_rate": 4.4071473158551814e-05,
|
| 541 |
+
"loss": 1.7238,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6008114856429463,
|
| 546 |
+
"grad_norm": 0.4091911315917969,
|
| 547 |
+
"learning_rate": 4.3993445692883896e-05,
|
| 548 |
+
"loss": 1.683,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6086142322097379,
|
| 553 |
+
"grad_norm": 0.38465380668640137,
|
| 554 |
+
"learning_rate": 4.3915418227215984e-05,
|
| 555 |
+
"loss": 1.6991,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6164169787765293,
|
| 560 |
+
"grad_norm": 0.4543341398239136,
|
| 561 |
+
"learning_rate": 4.383739076154807e-05,
|
| 562 |
+
"loss": 1.7105,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6242197253433208,
|
| 567 |
+
"grad_norm": 0.46507373452186584,
|
| 568 |
+
"learning_rate": 4.3759363295880154e-05,
|
| 569 |
+
"loss": 1.6935,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6320224719101124,
|
| 574 |
+
"grad_norm": 0.4509834349155426,
|
| 575 |
+
"learning_rate": 4.368133583021224e-05,
|
| 576 |
+
"loss": 1.6869,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6398252184769039,
|
| 581 |
+
"grad_norm": 0.5607530474662781,
|
| 582 |
+
"learning_rate": 4.3603308364544324e-05,
|
| 583 |
+
"loss": 1.7125,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6476279650436954,
|
| 588 |
+
"grad_norm": 0.4719931483268738,
|
| 589 |
+
"learning_rate": 4.3525280898876405e-05,
|
| 590 |
+
"loss": 1.677,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6554307116104869,
|
| 595 |
+
"grad_norm": 0.4882090091705322,
|
| 596 |
+
"learning_rate": 4.344725343320849e-05,
|
| 597 |
+
"loss": 1.6981,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6632334581772784,
|
| 602 |
+
"grad_norm": 0.5204262733459473,
|
| 603 |
+
"learning_rate": 4.3369225967540575e-05,
|
| 604 |
+
"loss": 1.6905,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6710362047440699,
|
| 609 |
+
"grad_norm": 0.5416198372840881,
|
| 610 |
+
"learning_rate": 4.329119850187266e-05,
|
| 611 |
+
"loss": 1.6874,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6788389513108615,
|
| 616 |
+
"grad_norm": 0.484465628862381,
|
| 617 |
+
"learning_rate": 4.3213171036204745e-05,
|
| 618 |
+
"loss": 1.6967,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.686641697877653,
|
| 623 |
+
"grad_norm": 0.4895637333393097,
|
| 624 |
+
"learning_rate": 4.313514357053683e-05,
|
| 625 |
+
"loss": 1.7088,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6944444444444444,
|
| 630 |
+
"grad_norm": 0.5524686574935913,
|
| 631 |
+
"learning_rate": 4.3057116104868915e-05,
|
| 632 |
+
"loss": 1.6826,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.702247191011236,
|
| 637 |
+
"grad_norm": 0.5131920576095581,
|
| 638 |
+
"learning_rate": 4.2979088639200997e-05,
|
| 639 |
+
"loss": 1.6932,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7100499375780275,
|
| 644 |
+
"grad_norm": 0.5526320934295654,
|
| 645 |
+
"learning_rate": 4.2901061173533085e-05,
|
| 646 |
+
"loss": 1.7017,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.717852684144819,
|
| 651 |
+
"grad_norm": 0.5150460600852966,
|
| 652 |
+
"learning_rate": 4.282303370786517e-05,
|
| 653 |
+
"loss": 1.6903,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7256554307116105,
|
| 658 |
+
"grad_norm": 0.5039018988609314,
|
| 659 |
+
"learning_rate": 4.2745006242197255e-05,
|
| 660 |
+
"loss": 1.7134,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.733458177278402,
|
| 665 |
+
"grad_norm": 0.48572325706481934,
|
| 666 |
+
"learning_rate": 4.266697877652934e-05,
|
| 667 |
+
"loss": 1.6976,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7412609238451935,
|
| 672 |
+
"grad_norm": 0.5142014026641846,
|
| 673 |
+
"learning_rate": 4.2588951310861425e-05,
|
| 674 |
+
"loss": 1.6834,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7490636704119851,
|
| 679 |
+
"grad_norm": 0.4606572985649109,
|
| 680 |
+
"learning_rate": 4.251092384519351e-05,
|
| 681 |
+
"loss": 1.6793,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7568664169787765,
|
| 686 |
+
"grad_norm": 0.46960020065307617,
|
| 687 |
+
"learning_rate": 4.2432896379525595e-05,
|
| 688 |
+
"loss": 1.6828,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.764669163545568,
|
| 693 |
+
"grad_norm": 0.4922361671924591,
|
| 694 |
+
"learning_rate": 4.235486891385768e-05,
|
| 695 |
+
"loss": 1.7022,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7724719101123596,
|
| 700 |
+
"grad_norm": 0.5289677381515503,
|
| 701 |
+
"learning_rate": 4.2276841448189764e-05,
|
| 702 |
+
"loss": 1.6903,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7802746566791511,
|
| 707 |
+
"grad_norm": 0.5616611838340759,
|
| 708 |
+
"learning_rate": 4.219881398252185e-05,
|
| 709 |
+
"loss": 1.6915,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7880774032459426,
|
| 714 |
+
"grad_norm": 0.4942910075187683,
|
| 715 |
+
"learning_rate": 4.2120786516853934e-05,
|
| 716 |
+
"loss": 1.6813,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.795880149812734,
|
| 721 |
+
"grad_norm": 0.5219690203666687,
|
| 722 |
+
"learning_rate": 4.204275905118602e-05,
|
| 723 |
+
"loss": 1.6731,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8036828963795256,
|
| 728 |
+
"grad_norm": 0.4913477897644043,
|
| 729 |
+
"learning_rate": 4.1964731585518104e-05,
|
| 730 |
+
"loss": 1.7001,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8114856429463171,
|
| 735 |
+
"grad_norm": 0.49623045325279236,
|
| 736 |
+
"learning_rate": 4.188670411985019e-05,
|
| 737 |
+
"loss": 1.686,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8192883895131086,
|
| 742 |
+
"grad_norm": 0.5660040974617004,
|
| 743 |
+
"learning_rate": 4.1808676654182274e-05,
|
| 744 |
+
"loss": 1.6794,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8270911360799001,
|
| 749 |
+
"grad_norm": 0.5747349262237549,
|
| 750 |
+
"learning_rate": 4.173064918851436e-05,
|
| 751 |
+
"loss": 1.6847,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8348938826466916,
|
| 756 |
+
"grad_norm": 0.5206372737884521,
|
| 757 |
+
"learning_rate": 4.1652621722846444e-05,
|
| 758 |
+
"loss": 1.6719,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8426966292134831,
|
| 763 |
+
"grad_norm": 0.5366120934486389,
|
| 764 |
+
"learning_rate": 4.1574594257178526e-05,
|
| 765 |
+
"loss": 1.6696,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8504993757802747,
|
| 770 |
+
"grad_norm": 0.6354568600654602,
|
| 771 |
+
"learning_rate": 4.1496566791510614e-05,
|
| 772 |
+
"loss": 1.6657,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8583021223470662,
|
| 777 |
+
"grad_norm": 0.5194640159606934,
|
| 778 |
+
"learning_rate": 4.1418539325842695e-05,
|
| 779 |
+
"loss": 1.6731,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8661048689138576,
|
| 784 |
+
"grad_norm": 0.43448275327682495,
|
| 785 |
+
"learning_rate": 4.1340511860174784e-05,
|
| 786 |
+
"loss": 1.6639,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8739076154806492,
|
| 791 |
+
"grad_norm": 0.5286650657653809,
|
| 792 |
+
"learning_rate": 4.1262484394506865e-05,
|
| 793 |
+
"loss": 1.677,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8817103620474407,
|
| 798 |
+
"grad_norm": 0.49659380316734314,
|
| 799 |
+
"learning_rate": 4.1184456928838954e-05,
|
| 800 |
+
"loss": 1.6758,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8895131086142322,
|
| 805 |
+
"grad_norm": 0.5224044322967529,
|
| 806 |
+
"learning_rate": 4.1106429463171035e-05,
|
| 807 |
+
"loss": 1.6528,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8973158551810237,
|
| 812 |
+
"grad_norm": 0.510977566242218,
|
| 813 |
+
"learning_rate": 4.1028401997503124e-05,
|
| 814 |
+
"loss": 1.6869,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9051186017478152,
|
| 819 |
+
"grad_norm": 0.5862101912498474,
|
| 820 |
+
"learning_rate": 4.0950374531835205e-05,
|
| 821 |
+
"loss": 1.68,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9129213483146067,
|
| 826 |
+
"grad_norm": 0.5646480321884155,
|
| 827 |
+
"learning_rate": 4.0872347066167293e-05,
|
| 828 |
+
"loss": 1.6749,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9207240948813983,
|
| 833 |
+
"grad_norm": 0.5872883200645447,
|
| 834 |
+
"learning_rate": 4.0794319600499375e-05,
|
| 835 |
+
"loss": 1.6661,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9285268414481898,
|
| 840 |
+
"grad_norm": 0.5308676958084106,
|
| 841 |
+
"learning_rate": 4.0716292134831463e-05,
|
| 842 |
+
"loss": 1.6747,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9363295880149812,
|
| 847 |
+
"grad_norm": 0.5872898101806641,
|
| 848 |
+
"learning_rate": 4.0638264669163545e-05,
|
| 849 |
+
"loss": 1.6719,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9441323345817728,
|
| 854 |
+
"grad_norm": 0.6066872477531433,
|
| 855 |
+
"learning_rate": 4.056023720349563e-05,
|
| 856 |
+
"loss": 1.6746,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.9519350811485643,
|
| 861 |
+
"grad_norm": 0.5329908132553101,
|
| 862 |
+
"learning_rate": 4.048220973782772e-05,
|
| 863 |
+
"loss": 1.6789,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9597378277153558,
|
| 868 |
+
"grad_norm": 0.4528316855430603,
|
| 869 |
+
"learning_rate": 4.04041822721598e-05,
|
| 870 |
+
"loss": 1.6767,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9675405742821473,
|
| 875 |
+
"grad_norm": 0.5394971966743469,
|
| 876 |
+
"learning_rate": 4.032615480649189e-05,
|
| 877 |
+
"loss": 1.6797,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9753433208489388,
|
| 882 |
+
"grad_norm": 0.6735771298408508,
|
| 883 |
+
"learning_rate": 4.024812734082397e-05,
|
| 884 |
+
"loss": 1.6842,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9831460674157303,
|
| 889 |
+
"grad_norm": 0.48064225912094116,
|
| 890 |
+
"learning_rate": 4.017009987515606e-05,
|
| 891 |
+
"loss": 1.6767,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.9909488139825219,
|
| 896 |
+
"grad_norm": 0.49285510182380676,
|
| 897 |
+
"learning_rate": 4.009207240948814e-05,
|
| 898 |
+
"loss": 1.666,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9987515605493134,
|
| 903 |
+
"grad_norm": 0.5762596726417542,
|
| 904 |
+
"learning_rate": 4.001404494382023e-05,
|
| 905 |
+
"loss": 1.6766,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.006554307116105,
|
| 910 |
+
"grad_norm": 0.4971337616443634,
|
| 911 |
+
"learning_rate": 3.993601747815231e-05,
|
| 912 |
+
"loss": 1.6727,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0143570536828963,
|
| 917 |
+
"grad_norm": 0.5485156178474426,
|
| 918 |
+
"learning_rate": 3.9857990012484394e-05,
|
| 919 |
+
"loss": 1.6842,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0221598002496879,
|
| 924 |
+
"grad_norm": 0.4900757968425751,
|
| 925 |
+
"learning_rate": 3.977996254681648e-05,
|
| 926 |
+
"loss": 1.6661,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.0299625468164795,
|
| 931 |
+
"grad_norm": 0.4844594895839691,
|
| 932 |
+
"learning_rate": 3.9701935081148564e-05,
|
| 933 |
+
"loss": 1.6601,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0377652933832708,
|
| 938 |
+
"grad_norm": 0.6316475868225098,
|
| 939 |
+
"learning_rate": 3.9623907615480646e-05,
|
| 940 |
+
"loss": 1.6481,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.0455680399500624,
|
| 945 |
+
"grad_norm": 0.6341625452041626,
|
| 946 |
+
"learning_rate": 3.9545880149812734e-05,
|
| 947 |
+
"loss": 1.6656,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.053370786516854,
|
| 952 |
+
"grad_norm": 0.6208726763725281,
|
| 953 |
+
"learning_rate": 3.9467852684144816e-05,
|
| 954 |
+
"loss": 1.6379,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.0611735330836454,
|
| 959 |
+
"grad_norm": 0.605767011642456,
|
| 960 |
+
"learning_rate": 3.9389825218476904e-05,
|
| 961 |
+
"loss": 1.6537,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.068976279650437,
|
| 966 |
+
"grad_norm": 0.5422509908676147,
|
| 967 |
+
"learning_rate": 3.931179775280899e-05,
|
| 968 |
+
"loss": 1.6689,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.0767790262172285,
|
| 973 |
+
"grad_norm": 0.6556758880615234,
|
| 974 |
+
"learning_rate": 3.9233770287141074e-05,
|
| 975 |
+
"loss": 1.6619,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.08458177278402,
|
| 980 |
+
"grad_norm": 0.5463916659355164,
|
| 981 |
+
"learning_rate": 3.915574282147316e-05,
|
| 982 |
+
"loss": 1.6584,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.0923845193508115,
|
| 987 |
+
"grad_norm": 0.660961925983429,
|
| 988 |
+
"learning_rate": 3.9077715355805244e-05,
|
| 989 |
+
"loss": 1.6813,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.100187265917603,
|
| 994 |
+
"grad_norm": 0.5504911541938782,
|
| 995 |
+
"learning_rate": 3.899968789013733e-05,
|
| 996 |
+
"loss": 1.6482,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1079900124843944,
|
| 1001 |
+
"grad_norm": 0.5077877044677734,
|
| 1002 |
+
"learning_rate": 3.8921660424469414e-05,
|
| 1003 |
+
"loss": 1.6508,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.115792759051186,
|
| 1008 |
+
"grad_norm": 0.5340275764465332,
|
| 1009 |
+
"learning_rate": 3.88436329588015e-05,
|
| 1010 |
+
"loss": 1.6685,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.1235955056179776,
|
| 1015 |
+
"grad_norm": 0.615564227104187,
|
| 1016 |
+
"learning_rate": 3.8765605493133584e-05,
|
| 1017 |
+
"loss": 1.6617,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.131398252184769,
|
| 1022 |
+
"grad_norm": 0.600592315196991,
|
| 1023 |
+
"learning_rate": 3.868757802746567e-05,
|
| 1024 |
+
"loss": 1.6478,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1392009987515606,
|
| 1029 |
+
"grad_norm": 0.606829047203064,
|
| 1030 |
+
"learning_rate": 3.8609550561797754e-05,
|
| 1031 |
+
"loss": 1.6627,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1470037453183521,
|
| 1036 |
+
"grad_norm": 0.5715992450714111,
|
| 1037 |
+
"learning_rate": 3.853152309612984e-05,
|
| 1038 |
+
"loss": 1.6724,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.1548064918851435,
|
| 1043 |
+
"grad_norm": 0.5475966334342957,
|
| 1044 |
+
"learning_rate": 3.8453495630461923e-05,
|
| 1045 |
+
"loss": 1.6762,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.162609238451935,
|
| 1050 |
+
"grad_norm": 0.5486684441566467,
|
| 1051 |
+
"learning_rate": 3.837546816479401e-05,
|
| 1052 |
+
"loss": 1.6418,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.1704119850187267,
|
| 1057 |
+
"grad_norm": 0.5656526684761047,
|
| 1058 |
+
"learning_rate": 3.829744069912609e-05,
|
| 1059 |
+
"loss": 1.6623,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.178214731585518,
|
| 1064 |
+
"grad_norm": 0.471967875957489,
|
| 1065 |
+
"learning_rate": 3.821941323345818e-05,
|
| 1066 |
+
"loss": 1.6339,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.1860174781523096,
|
| 1071 |
+
"grad_norm": 0.5814192891120911,
|
| 1072 |
+
"learning_rate": 3.814138576779026e-05,
|
| 1073 |
+
"loss": 1.6626,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.1938202247191012,
|
| 1078 |
+
"grad_norm": 0.5809513926506042,
|
| 1079 |
+
"learning_rate": 3.806335830212235e-05,
|
| 1080 |
+
"loss": 1.6521,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2016229712858926,
|
| 1085 |
+
"grad_norm": 0.564431369304657,
|
| 1086 |
+
"learning_rate": 3.798533083645443e-05,
|
| 1087 |
+
"loss": 1.633,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2094257178526842,
|
| 1092 |
+
"grad_norm": 0.5864349007606506,
|
| 1093 |
+
"learning_rate": 3.7907303370786515e-05,
|
| 1094 |
+
"loss": 1.6724,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.2172284644194757,
|
| 1099 |
+
"grad_norm": 0.5883368849754333,
|
| 1100 |
+
"learning_rate": 3.78292759051186e-05,
|
| 1101 |
+
"loss": 1.6661,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.225031210986267,
|
| 1106 |
+
"grad_norm": 0.5318378806114197,
|
| 1107 |
+
"learning_rate": 3.7751248439450685e-05,
|
| 1108 |
+
"loss": 1.6703,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2328339575530587,
|
| 1113 |
+
"grad_norm": 0.6735764741897583,
|
| 1114 |
+
"learning_rate": 3.767322097378277e-05,
|
| 1115 |
+
"loss": 1.6558,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2406367041198503,
|
| 1120 |
+
"grad_norm": 0.5900487303733826,
|
| 1121 |
+
"learning_rate": 3.7595193508114855e-05,
|
| 1122 |
+
"loss": 1.6472,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.2484394506866416,
|
| 1127 |
+
"grad_norm": 0.4971151649951935,
|
| 1128 |
+
"learning_rate": 3.751716604244694e-05,
|
| 1129 |
+
"loss": 1.6357,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2562421972534332,
|
| 1134 |
+
"grad_norm": 0.6045508980751038,
|
| 1135 |
+
"learning_rate": 3.7439138576779024e-05,
|
| 1136 |
+
"loss": 1.6556,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2640449438202248,
|
| 1141 |
+
"grad_norm": 0.5860553979873657,
|
| 1142 |
+
"learning_rate": 3.736111111111111e-05,
|
| 1143 |
+
"loss": 1.6659,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2718476903870162,
|
| 1148 |
+
"grad_norm": 0.5339462161064148,
|
| 1149 |
+
"learning_rate": 3.7283083645443194e-05,
|
| 1150 |
+
"loss": 1.6707,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.2796504369538078,
|
| 1155 |
+
"grad_norm": 0.5763932466506958,
|
| 1156 |
+
"learning_rate": 3.720505617977528e-05,
|
| 1157 |
+
"loss": 1.6694,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.2874531835205993,
|
| 1162 |
+
"grad_norm": 0.5927013754844666,
|
| 1163 |
+
"learning_rate": 3.712702871410737e-05,
|
| 1164 |
+
"loss": 1.6258,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.2952559300873907,
|
| 1169 |
+
"grad_norm": 0.6203845739364624,
|
| 1170 |
+
"learning_rate": 3.704900124843945e-05,
|
| 1171 |
+
"loss": 1.6459,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3030586766541823,
|
| 1176 |
+
"grad_norm": 0.5438473224639893,
|
| 1177 |
+
"learning_rate": 3.697097378277154e-05,
|
| 1178 |
+
"loss": 1.6352,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.3108614232209739,
|
| 1183 |
+
"grad_norm": 0.6493474245071411,
|
| 1184 |
+
"learning_rate": 3.689294631710362e-05,
|
| 1185 |
+
"loss": 1.6643,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3186641697877652,
|
| 1190 |
+
"grad_norm": 0.5788607597351074,
|
| 1191 |
+
"learning_rate": 3.681491885143571e-05,
|
| 1192 |
+
"loss": 1.6753,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3264669163545568,
|
| 1197 |
+
"grad_norm": 0.5591830015182495,
|
| 1198 |
+
"learning_rate": 3.673689138576779e-05,
|
| 1199 |
+
"loss": 1.6316,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3342696629213484,
|
| 1204 |
+
"grad_norm": 0.4842735230922699,
|
| 1205 |
+
"learning_rate": 3.665886392009988e-05,
|
| 1206 |
+
"loss": 1.6699,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.3420724094881398,
|
| 1211 |
+
"grad_norm": 0.6692916750907898,
|
| 1212 |
+
"learning_rate": 3.658083645443196e-05,
|
| 1213 |
+
"loss": 1.6249,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.3498751560549314,
|
| 1218 |
+
"grad_norm": 0.5876237154006958,
|
| 1219 |
+
"learning_rate": 3.650280898876405e-05,
|
| 1220 |
+
"loss": 1.6576,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.357677902621723,
|
| 1225 |
+
"grad_norm": 0.6215786933898926,
|
| 1226 |
+
"learning_rate": 3.642478152309613e-05,
|
| 1227 |
+
"loss": 1.6332,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.3654806491885143,
|
| 1232 |
+
"grad_norm": 0.54453444480896,
|
| 1233 |
+
"learning_rate": 3.634675405742822e-05,
|
| 1234 |
+
"loss": 1.6456,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.373283395755306,
|
| 1239 |
+
"grad_norm": 0.6348562836647034,
|
| 1240 |
+
"learning_rate": 3.62687265917603e-05,
|
| 1241 |
+
"loss": 1.6476,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.3810861423220975,
|
| 1246 |
+
"grad_norm": 0.6236295700073242,
|
| 1247 |
+
"learning_rate": 3.6190699126092384e-05,
|
| 1248 |
+
"loss": 1.6676,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.3888888888888888,
|
| 1253 |
+
"grad_norm": 0.5054611563682556,
|
| 1254 |
+
"learning_rate": 3.611267166042447e-05,
|
| 1255 |
+
"loss": 1.6506,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.3966916354556804,
|
| 1260 |
+
"grad_norm": 0.582488477230072,
|
| 1261 |
+
"learning_rate": 3.6034644194756553e-05,
|
| 1262 |
+
"loss": 1.6413,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.404494382022472,
|
| 1267 |
+
"grad_norm": 0.6589245796203613,
|
| 1268 |
+
"learning_rate": 3.595661672908864e-05,
|
| 1269 |
+
"loss": 1.6824,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.4122971285892634,
|
| 1274 |
+
"grad_norm": 0.5584797859191895,
|
| 1275 |
+
"learning_rate": 3.587858926342072e-05,
|
| 1276 |
+
"loss": 1.6685,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.420099875156055,
|
| 1281 |
+
"grad_norm": 0.6652534604072571,
|
| 1282 |
+
"learning_rate": 3.580056179775281e-05,
|
| 1283 |
+
"loss": 1.6362,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.4279026217228465,
|
| 1288 |
+
"grad_norm": 0.6449257731437683,
|
| 1289 |
+
"learning_rate": 3.572253433208489e-05,
|
| 1290 |
+
"loss": 1.6543,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.435705368289638,
|
| 1295 |
+
"grad_norm": 0.6358399391174316,
|
| 1296 |
+
"learning_rate": 3.564450686641698e-05,
|
| 1297 |
+
"loss": 1.6464,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.4435081148564295,
|
| 1302 |
+
"grad_norm": 0.6031101942062378,
|
| 1303 |
+
"learning_rate": 3.556647940074906e-05,
|
| 1304 |
+
"loss": 1.6435,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.451310861423221,
|
| 1309 |
+
"grad_norm": 0.5363774299621582,
|
| 1310 |
+
"learning_rate": 3.548845193508115e-05,
|
| 1311 |
+
"loss": 1.6728,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.4591136079900124,
|
| 1316 |
+
"grad_norm": 0.6634340286254883,
|
| 1317 |
+
"learning_rate": 3.541042446941323e-05,
|
| 1318 |
+
"loss": 1.6638,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.466916354556804,
|
| 1323 |
+
"grad_norm": 0.6200147867202759,
|
| 1324 |
+
"learning_rate": 3.533239700374532e-05,
|
| 1325 |
+
"loss": 1.6537,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.4747191011235956,
|
| 1330 |
+
"grad_norm": 0.5800793766975403,
|
| 1331 |
+
"learning_rate": 3.52543695380774e-05,
|
| 1332 |
+
"loss": 1.6549,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.482521847690387,
|
| 1337 |
+
"grad_norm": 0.5839795470237732,
|
| 1338 |
+
"learning_rate": 3.517634207240949e-05,
|
| 1339 |
+
"loss": 1.6571,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.4903245942571786,
|
| 1344 |
+
"grad_norm": 0.5768577456474304,
|
| 1345 |
+
"learning_rate": 3.509831460674157e-05,
|
| 1346 |
+
"loss": 1.679,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.4981273408239701,
|
| 1351 |
+
"grad_norm": 0.5268595218658447,
|
| 1352 |
+
"learning_rate": 3.502028714107366e-05,
|
| 1353 |
+
"loss": 1.6461,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.5059300873907615,
|
| 1358 |
+
"grad_norm": 0.6356619000434875,
|
| 1359 |
+
"learning_rate": 3.494225967540574e-05,
|
| 1360 |
+
"loss": 1.6312,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.513732833957553,
|
| 1365 |
+
"grad_norm": 0.5722295641899109,
|
| 1366 |
+
"learning_rate": 3.486423220973783e-05,
|
| 1367 |
+
"loss": 1.6418,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.5215355805243447,
|
| 1372 |
+
"grad_norm": 0.5974990129470825,
|
| 1373 |
+
"learning_rate": 3.478620474406992e-05,
|
| 1374 |
+
"loss": 1.6526,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.529338327091136,
|
| 1379 |
+
"grad_norm": 0.6584164500236511,
|
| 1380 |
+
"learning_rate": 3.4708177278402e-05,
|
| 1381 |
+
"loss": 1.6749,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.5371410736579276,
|
| 1386 |
+
"grad_norm": 0.6195454001426697,
|
| 1387 |
+
"learning_rate": 3.463014981273409e-05,
|
| 1388 |
+
"loss": 1.6406,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.5449438202247192,
|
| 1393 |
+
"grad_norm": 0.5923195481300354,
|
| 1394 |
+
"learning_rate": 3.455212234706617e-05,
|
| 1395 |
+
"loss": 1.6471,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.5527465667915106,
|
| 1400 |
+
"grad_norm": 0.59232097864151,
|
| 1401 |
+
"learning_rate": 3.447409488139825e-05,
|
| 1402 |
+
"loss": 1.6595,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.5605493133583022,
|
| 1407 |
+
"grad_norm": 0.5838867425918579,
|
| 1408 |
+
"learning_rate": 3.439606741573034e-05,
|
| 1409 |
+
"loss": 1.6449,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.5683520599250937,
|
| 1414 |
+
"grad_norm": 0.6070720553398132,
|
| 1415 |
+
"learning_rate": 3.431803995006242e-05,
|
| 1416 |
+
"loss": 1.6376,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.576154806491885,
|
| 1421 |
+
"grad_norm": 0.5864161849021912,
|
| 1422 |
+
"learning_rate": 3.4240012484394504e-05,
|
| 1423 |
+
"loss": 1.6555,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.5839575530586767,
|
| 1428 |
+
"grad_norm": 0.6388084292411804,
|
| 1429 |
+
"learning_rate": 3.416198501872659e-05,
|
| 1430 |
+
"loss": 1.6522,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.5917602996254683,
|
| 1435 |
+
"grad_norm": 0.5700277090072632,
|
| 1436 |
+
"learning_rate": 3.4083957553058674e-05,
|
| 1437 |
+
"loss": 1.6347,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.5995630461922596,
|
| 1442 |
+
"grad_norm": 0.6094324588775635,
|
| 1443 |
+
"learning_rate": 3.400593008739076e-05,
|
| 1444 |
+
"loss": 1.6439,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.6073657927590512,
|
| 1449 |
+
"grad_norm": 0.6227761507034302,
|
| 1450 |
+
"learning_rate": 3.3927902621722844e-05,
|
| 1451 |
+
"loss": 1.6567,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.6151685393258428,
|
| 1456 |
+
"grad_norm": 0.6303547024726868,
|
| 1457 |
+
"learning_rate": 3.384987515605493e-05,
|
| 1458 |
+
"loss": 1.6221,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.6229712858926342,
|
| 1463 |
+
"grad_norm": 0.6025314927101135,
|
| 1464 |
+
"learning_rate": 3.3771847690387014e-05,
|
| 1465 |
+
"loss": 1.6371,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.6307740324594258,
|
| 1470 |
+
"grad_norm": 0.683813214302063,
|
| 1471 |
+
"learning_rate": 3.36938202247191e-05,
|
| 1472 |
+
"loss": 1.6364,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6385767790262173,
|
| 1477 |
+
"grad_norm": 0.763845682144165,
|
| 1478 |
+
"learning_rate": 3.361579275905119e-05,
|
| 1479 |
+
"loss": 1.6274,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6463795255930087,
|
| 1484 |
+
"grad_norm": 0.6550936698913574,
|
| 1485 |
+
"learning_rate": 3.353776529338327e-05,
|
| 1486 |
+
"loss": 1.6433,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.6541822721598003,
|
| 1491 |
+
"grad_norm": 0.6213018894195557,
|
| 1492 |
+
"learning_rate": 3.345973782771536e-05,
|
| 1493 |
+
"loss": 1.6405,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.6619850187265919,
|
| 1498 |
+
"grad_norm": 0.6199821829795837,
|
| 1499 |
+
"learning_rate": 3.338171036204744e-05,
|
| 1500 |
+
"loss": 1.6221,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.6697877652933832,
|
| 1505 |
+
"grad_norm": 0.6940792798995972,
|
| 1506 |
+
"learning_rate": 3.330368289637953e-05,
|
| 1507 |
+
"loss": 1.6064,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.6775905118601748,
|
| 1512 |
+
"grad_norm": 0.5895411968231201,
|
| 1513 |
+
"learning_rate": 3.322565543071161e-05,
|
| 1514 |
+
"loss": 1.6246,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.6853932584269664,
|
| 1519 |
+
"grad_norm": 0.584697425365448,
|
| 1520 |
+
"learning_rate": 3.31476279650437e-05,
|
| 1521 |
+
"loss": 1.6164,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.6931960049937578,
|
| 1526 |
+
"grad_norm": 0.5995935201644897,
|
| 1527 |
+
"learning_rate": 3.306960049937578e-05,
|
| 1528 |
+
"loss": 1.6303,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.7009987515605494,
|
| 1533 |
+
"grad_norm": 0.6184208989143372,
|
| 1534 |
+
"learning_rate": 3.299157303370787e-05,
|
| 1535 |
+
"loss": 1.6226,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.708801498127341,
|
| 1540 |
+
"grad_norm": 0.6035963892936707,
|
| 1541 |
+
"learning_rate": 3.291354556803995e-05,
|
| 1542 |
+
"loss": 1.6519,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.7166042446941323,
|
| 1547 |
+
"grad_norm": 0.6514495611190796,
|
| 1548 |
+
"learning_rate": 3.283551810237204e-05,
|
| 1549 |
+
"loss": 1.6311,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.724406991260924,
|
| 1554 |
+
"grad_norm": 0.681898832321167,
|
| 1555 |
+
"learning_rate": 3.275749063670412e-05,
|
| 1556 |
+
"loss": 1.6493,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.7322097378277155,
|
| 1561 |
+
"grad_norm": 0.641394317150116,
|
| 1562 |
+
"learning_rate": 3.267946317103621e-05,
|
| 1563 |
+
"loss": 1.6562,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.7400124843945068,
|
| 1568 |
+
"grad_norm": 0.6565654277801514,
|
| 1569 |
+
"learning_rate": 3.260143570536829e-05,
|
| 1570 |
+
"loss": 1.6193,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.7478152309612984,
|
| 1575 |
+
"grad_norm": 0.6501032114028931,
|
| 1576 |
+
"learning_rate": 3.252340823970037e-05,
|
| 1577 |
+
"loss": 1.6294,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.75561797752809,
|
| 1582 |
+
"grad_norm": 0.5766665935516357,
|
| 1583 |
+
"learning_rate": 3.244538077403246e-05,
|
| 1584 |
+
"loss": 1.6236,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.7634207240948814,
|
| 1589 |
+
"grad_norm": 0.6376942992210388,
|
| 1590 |
+
"learning_rate": 3.236735330836454e-05,
|
| 1591 |
+
"loss": 1.6592,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.771223470661673,
|
| 1596 |
+
"grad_norm": 0.6972282528877258,
|
| 1597 |
+
"learning_rate": 3.228932584269663e-05,
|
| 1598 |
+
"loss": 1.6173,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.7790262172284645,
|
| 1603 |
+
"grad_norm": 0.6171312928199768,
|
| 1604 |
+
"learning_rate": 3.221129837702871e-05,
|
| 1605 |
+
"loss": 1.6414,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.786828963795256,
|
| 1610 |
+
"grad_norm": 0.715184211730957,
|
| 1611 |
+
"learning_rate": 3.21332709113608e-05,
|
| 1612 |
+
"loss": 1.637,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.7946317103620475,
|
| 1617 |
+
"grad_norm": 0.6628192663192749,
|
| 1618 |
+
"learning_rate": 3.205524344569288e-05,
|
| 1619 |
+
"loss": 1.6102,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.802434456928839,
|
| 1624 |
+
"grad_norm": 0.6058287024497986,
|
| 1625 |
+
"learning_rate": 3.197721598002497e-05,
|
| 1626 |
+
"loss": 1.6512,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.8102372034956304,
|
| 1631 |
+
"grad_norm": 0.6275887489318848,
|
| 1632 |
+
"learning_rate": 3.189918851435705e-05,
|
| 1633 |
+
"loss": 1.6435,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.818039950062422,
|
| 1638 |
+
"grad_norm": 0.7389242053031921,
|
| 1639 |
+
"learning_rate": 3.182116104868914e-05,
|
| 1640 |
+
"loss": 1.6234,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.8258426966292136,
|
| 1645 |
+
"grad_norm": 0.649131715297699,
|
| 1646 |
+
"learning_rate": 3.174313358302122e-05,
|
| 1647 |
+
"loss": 1.6332,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.833645443196005,
|
| 1652 |
+
"grad_norm": 0.5898476839065552,
|
| 1653 |
+
"learning_rate": 3.166510611735331e-05,
|
| 1654 |
+
"loss": 1.6299,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.8414481897627963,
|
| 1659 |
+
"grad_norm": 0.617365837097168,
|
| 1660 |
+
"learning_rate": 3.158707865168539e-05,
|
| 1661 |
+
"loss": 1.608,
|
| 1662 |
+
"step": 11800
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.8492509363295881,
|
| 1666 |
+
"grad_norm": 0.6347021460533142,
|
| 1667 |
+
"learning_rate": 3.150905118601748e-05,
|
| 1668 |
+
"loss": 1.6168,
|
| 1669 |
+
"step": 11850
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.8570536828963795,
|
| 1673 |
+
"grad_norm": 0.6479565501213074,
|
| 1674 |
+
"learning_rate": 3.143102372034957e-05,
|
| 1675 |
+
"loss": 1.6089,
|
| 1676 |
+
"step": 11900
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.8648564294631709,
|
| 1680 |
+
"grad_norm": 0.6168213486671448,
|
| 1681 |
+
"learning_rate": 3.135299625468165e-05,
|
| 1682 |
+
"loss": 1.63,
|
| 1683 |
+
"step": 11950
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.8726591760299627,
|
| 1687 |
+
"grad_norm": 0.5773766040802002,
|
| 1688 |
+
"learning_rate": 3.127496878901374e-05,
|
| 1689 |
+
"loss": 1.6183,
|
| 1690 |
+
"step": 12000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.880461922596754,
|
| 1694 |
+
"grad_norm": 0.5600804686546326,
|
| 1695 |
+
"learning_rate": 3.119694132334582e-05,
|
| 1696 |
+
"loss": 1.6393,
|
| 1697 |
+
"step": 12050
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.8882646691635454,
|
| 1701 |
+
"grad_norm": 0.623058557510376,
|
| 1702 |
+
"learning_rate": 3.111891385767791e-05,
|
| 1703 |
+
"loss": 1.6112,
|
| 1704 |
+
"step": 12100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.8960674157303372,
|
| 1708 |
+
"grad_norm": 0.5952323079109192,
|
| 1709 |
+
"learning_rate": 3.104088639200999e-05,
|
| 1710 |
+
"loss": 1.6319,
|
| 1711 |
+
"step": 12150
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.9038701622971286,
|
| 1715 |
+
"grad_norm": 0.7055174112319946,
|
| 1716 |
+
"learning_rate": 3.096285892634208e-05,
|
| 1717 |
+
"loss": 1.6275,
|
| 1718 |
+
"step": 12200
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.91167290886392,
|
| 1722 |
+
"grad_norm": 0.5625096559524536,
|
| 1723 |
+
"learning_rate": 3.088483146067416e-05,
|
| 1724 |
+
"loss": 1.6345,
|
| 1725 |
+
"step": 12250
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.9194756554307117,
|
| 1729 |
+
"grad_norm": 0.5937293767929077,
|
| 1730 |
+
"learning_rate": 3.080680399500624e-05,
|
| 1731 |
+
"loss": 1.6197,
|
| 1732 |
+
"step": 12300
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.927278401997503,
|
| 1736 |
+
"grad_norm": 0.6606655120849609,
|
| 1737 |
+
"learning_rate": 3.072877652933833e-05,
|
| 1738 |
+
"loss": 1.6201,
|
| 1739 |
+
"step": 12350
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.9350811485642945,
|
| 1743 |
+
"grad_norm": 0.6392807960510254,
|
| 1744 |
+
"learning_rate": 3.065074906367041e-05,
|
| 1745 |
+
"loss": 1.6309,
|
| 1746 |
+
"step": 12400
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.9428838951310863,
|
| 1750 |
+
"grad_norm": 0.7471784353256226,
|
| 1751 |
+
"learning_rate": 3.057272159800249e-05,
|
| 1752 |
+
"loss": 1.6131,
|
| 1753 |
+
"step": 12450
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.9506866416978776,
|
| 1757 |
+
"grad_norm": 0.6735255718231201,
|
| 1758 |
+
"learning_rate": 3.0494694132334585e-05,
|
| 1759 |
+
"loss": 1.6514,
|
| 1760 |
+
"step": 12500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.958489388264669,
|
| 1764 |
+
"grad_norm": 0.6417968273162842,
|
| 1765 |
+
"learning_rate": 3.0416666666666666e-05,
|
| 1766 |
+
"loss": 1.6245,
|
| 1767 |
+
"step": 12550
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.9662921348314608,
|
| 1771 |
+
"grad_norm": 0.6633313894271851,
|
| 1772 |
+
"learning_rate": 3.0338639200998755e-05,
|
| 1773 |
+
"loss": 1.6267,
|
| 1774 |
+
"step": 12600
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.9740948813982522,
|
| 1778 |
+
"grad_norm": 0.6218631267547607,
|
| 1779 |
+
"learning_rate": 3.026061173533084e-05,
|
| 1780 |
+
"loss": 1.6472,
|
| 1781 |
+
"step": 12650
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.9818976279650435,
|
| 1785 |
+
"grad_norm": 0.594956636428833,
|
| 1786 |
+
"learning_rate": 3.018258426966292e-05,
|
| 1787 |
+
"loss": 1.6298,
|
| 1788 |
+
"step": 12700
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.9897003745318353,
|
| 1792 |
+
"grad_norm": 0.5888795852661133,
|
| 1793 |
+
"learning_rate": 3.010455680399501e-05,
|
| 1794 |
+
"loss": 1.626,
|
| 1795 |
+
"step": 12750
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.9975031210986267,
|
| 1799 |
+
"grad_norm": 0.6327818632125854,
|
| 1800 |
+
"learning_rate": 3.002652933832709e-05,
|
| 1801 |
+
"loss": 1.625,
|
| 1802 |
+
"step": 12800
|
| 1803 |
+
}
|
| 1804 |
+
],
|
| 1805 |
+
"logging_steps": 50,
|
| 1806 |
+
"max_steps": 32040,
|
| 1807 |
+
"num_input_tokens_seen": 0,
|
| 1808 |
+
"num_train_epochs": 5,
|
| 1809 |
+
"save_steps": 100,
|
| 1810 |
+
"stateful_callbacks": {
|
| 1811 |
+
"TrainerControl": {
|
| 1812 |
+
"args": {
|
| 1813 |
+
"should_epoch_stop": false,
|
| 1814 |
+
"should_evaluate": false,
|
| 1815 |
+
"should_log": false,
|
| 1816 |
+
"should_save": true,
|
| 1817 |
+
"should_training_stop": false
|
| 1818 |
+
},
|
| 1819 |
+
"attributes": {}
|
| 1820 |
+
}
|
| 1821 |
+
},
|
| 1822 |
+
"total_flos": 1.7153416265151283e+17,
|
| 1823 |
+
"train_batch_size": 10,
|
| 1824 |
+
"trial_name": null,
|
| 1825 |
+
"trial_params": null
|
| 1826 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:333d15756d4fa369e43e1866c86b7100a6f132939cab12ac6cf75e7de5b1d188
|
| 3 |
+
size 5304
|