Training in progress, epoch 20, checkpoint
Browse files- checkpoint-170/config.json +27 -0
- checkpoint-170/generation_config.json +6 -0
- checkpoint-170/model.safetensors +3 -0
- checkpoint-170/optimizer.pt +3 -0
- checkpoint-170/rng_state.pth +3 -0
- checkpoint-170/scheduler.pt +3 -0
- checkpoint-170/special_tokens_map.json +24 -0
- checkpoint-170/tokenizer.json +0 -0
- checkpoint-170/tokenizer_config.json +78 -0
- checkpoint-170/trainer_state.json +1039 -0
- checkpoint-170/training_args.bin +3 -0
checkpoint-170/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lectura/TinyLlama-120M-news",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 2048,
|
| 13 |
+
"max_position_embeddings": 1024,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"num_hidden_layers": 12,
|
| 17 |
+
"num_key_value_heads": 12,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": true,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.35.0",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 46336
|
| 27 |
+
}
|
checkpoint-170/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.35.0"
|
| 6 |
+
}
|
checkpoint-170/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7b31abefafd3b7609b0d76d8d5f5c15dbccd4f96772baabe021b006e4727656
|
| 3 |
+
size 482171848
|
checkpoint-170/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae3461a050cfa53960a8e617f7b7b5d5837542afe70d362d1b5733842a145fc0
|
| 3 |
+
size 964412154
|
checkpoint-170/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d2633b30d6f8bc95e53ac59f43880033523b63ea88408f92c1df3391a619329
|
| 3 |
+
size 14244
|
checkpoint-170/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21e8550c89a60068edadd26373cfbd6774a64cc9105920299d12d952bdf04bb9
|
| 3 |
+
size 1064
|
checkpoint-170/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": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-170/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-170/tokenizer_config.json
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"46331": {
|
| 28 |
+
"content": "<|sep|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"46332": {
|
| 36 |
+
"content": "<|endoftext|>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"46333": {
|
| 44 |
+
"content": "<|acc|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"46334": {
|
| 52 |
+
"content": "<|rrn|>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"46335": {
|
| 60 |
+
"content": "<|tel|>",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
}
|
| 67 |
+
},
|
| 68 |
+
"bos_token": "<s>",
|
| 69 |
+
"clean_up_tokenization_spaces": false,
|
| 70 |
+
"eos_token": "</s>",
|
| 71 |
+
"legacy": false,
|
| 72 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 73 |
+
"pad_token": "<unk>",
|
| 74 |
+
"sp_model_kwargs": {},
|
| 75 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 76 |
+
"unk_token": "<unk>",
|
| 77 |
+
"use_default_system_prompt": false
|
| 78 |
+
}
|
checkpoint-170/trainer_state.json
ADDED
|
@@ -0,0 +1,1039 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.923076923076923,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 170,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.12,
|
| 13 |
+
"learning_rate": 9.975000000000002e-06,
|
| 14 |
+
"loss": 12.7837,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.25,
|
| 19 |
+
"learning_rate": 9.950000000000001e-06,
|
| 20 |
+
"loss": 12.4595,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.37,
|
| 25 |
+
"learning_rate": 9.925e-06,
|
| 26 |
+
"loss": 11.9117,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.49,
|
| 31 |
+
"learning_rate": 9.9e-06,
|
| 32 |
+
"loss": 11.662,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.62,
|
| 37 |
+
"learning_rate": 9.875000000000001e-06,
|
| 38 |
+
"loss": 11.4278,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.74,
|
| 43 |
+
"learning_rate": 9.85e-06,
|
| 44 |
+
"loss": 11.1398,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.86,
|
| 49 |
+
"learning_rate": 9.825000000000002e-06,
|
| 50 |
+
"loss": 10.9135,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.98,
|
| 55 |
+
"learning_rate": 9.800000000000001e-06,
|
| 56 |
+
"loss": 10.8274,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.11,
|
| 61 |
+
"learning_rate": 9.775e-06,
|
| 62 |
+
"loss": 10.5068,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.23,
|
| 67 |
+
"learning_rate": 9.75e-06,
|
| 68 |
+
"loss": 10.3551,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.35,
|
| 73 |
+
"learning_rate": 9.725000000000001e-06,
|
| 74 |
+
"loss": 10.1776,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.48,
|
| 79 |
+
"learning_rate": 9.7e-06,
|
| 80 |
+
"loss": 10.0708,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.6,
|
| 85 |
+
"learning_rate": 9.675000000000001e-06,
|
| 86 |
+
"loss": 9.8703,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.72,
|
| 91 |
+
"learning_rate": 9.65e-06,
|
| 92 |
+
"loss": 9.7534,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.85,
|
| 97 |
+
"learning_rate": 9.625e-06,
|
| 98 |
+
"loss": 9.6094,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.97,
|
| 103 |
+
"learning_rate": 9.600000000000001e-06,
|
| 104 |
+
"loss": 9.3989,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"learning_rate": 9.575e-06,
|
| 110 |
+
"loss": 9.2753,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.22,
|
| 115 |
+
"learning_rate": 9.55e-06,
|
| 116 |
+
"loss": 9.138,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 2.34,
|
| 121 |
+
"learning_rate": 9.525000000000001e-06,
|
| 122 |
+
"loss": 9.0598,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.46,
|
| 127 |
+
"learning_rate": 9.5e-06,
|
| 128 |
+
"loss": 8.8535,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.58,
|
| 133 |
+
"learning_rate": 9.475000000000002e-06,
|
| 134 |
+
"loss": 8.7378,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.71,
|
| 139 |
+
"learning_rate": 9.450000000000001e-06,
|
| 140 |
+
"loss": 8.6082,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.83,
|
| 145 |
+
"learning_rate": 9.425e-06,
|
| 146 |
+
"loss": 8.4904,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.95,
|
| 151 |
+
"learning_rate": 9.4e-06,
|
| 152 |
+
"loss": 8.2918,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.08,
|
| 157 |
+
"learning_rate": 9.375000000000001e-06,
|
| 158 |
+
"loss": 8.2553,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.2,
|
| 163 |
+
"learning_rate": 9.350000000000002e-06,
|
| 164 |
+
"loss": 8.0724,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.32,
|
| 169 |
+
"learning_rate": 9.325000000000001e-06,
|
| 170 |
+
"loss": 8.0029,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 3.45,
|
| 175 |
+
"learning_rate": 9.3e-06,
|
| 176 |
+
"loss": 7.8227,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.57,
|
| 181 |
+
"learning_rate": 9.275e-06,
|
| 182 |
+
"loss": 7.7003,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.69,
|
| 187 |
+
"learning_rate": 9.250000000000001e-06,
|
| 188 |
+
"loss": 7.5943,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.82,
|
| 193 |
+
"learning_rate": 9.225e-06,
|
| 194 |
+
"loss": 7.5119,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.94,
|
| 199 |
+
"learning_rate": 9.200000000000002e-06,
|
| 200 |
+
"loss": 7.3673,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 4.06,
|
| 205 |
+
"learning_rate": 9.175000000000001e-06,
|
| 206 |
+
"loss": 7.3051,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 4.18,
|
| 211 |
+
"learning_rate": 9.15e-06,
|
| 212 |
+
"loss": 7.224,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.31,
|
| 217 |
+
"learning_rate": 9.125e-06,
|
| 218 |
+
"loss": 7.0517,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 4.43,
|
| 223 |
+
"learning_rate": 9.100000000000001e-06,
|
| 224 |
+
"loss": 7.0021,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 4.55,
|
| 229 |
+
"learning_rate": 9.075e-06,
|
| 230 |
+
"loss": 6.9407,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 4.68,
|
| 235 |
+
"learning_rate": 9.050000000000001e-06,
|
| 236 |
+
"loss": 6.7762,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.8,
|
| 241 |
+
"learning_rate": 9.025e-06,
|
| 242 |
+
"loss": 6.7675,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.92,
|
| 247 |
+
"learning_rate": 9e-06,
|
| 248 |
+
"loss": 6.6573,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 5.05,
|
| 253 |
+
"learning_rate": 8.975e-06,
|
| 254 |
+
"loss": 6.6093,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.17,
|
| 259 |
+
"learning_rate": 8.95e-06,
|
| 260 |
+
"loss": 6.5268,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 5.29,
|
| 265 |
+
"learning_rate": 8.925e-06,
|
| 266 |
+
"loss": 6.4456,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 5.42,
|
| 271 |
+
"learning_rate": 8.900000000000001e-06,
|
| 272 |
+
"loss": 6.4134,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 5.54,
|
| 277 |
+
"learning_rate": 8.875e-06,
|
| 278 |
+
"loss": 6.3414,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 5.66,
|
| 283 |
+
"learning_rate": 8.85e-06,
|
| 284 |
+
"loss": 6.2924,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 5.78,
|
| 289 |
+
"learning_rate": 8.825000000000001e-06,
|
| 290 |
+
"loss": 6.2176,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 5.91,
|
| 295 |
+
"learning_rate": 8.8e-06,
|
| 296 |
+
"loss": 6.1821,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 6.03,
|
| 301 |
+
"learning_rate": 8.775e-06,
|
| 302 |
+
"loss": 6.1344,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 6.15,
|
| 307 |
+
"learning_rate": 8.750000000000001e-06,
|
| 308 |
+
"loss": 6.0681,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 6.28,
|
| 313 |
+
"learning_rate": 8.725000000000002e-06,
|
| 314 |
+
"loss": 6.0286,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.4,
|
| 319 |
+
"learning_rate": 8.700000000000001e-06,
|
| 320 |
+
"loss": 5.9494,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 6.52,
|
| 325 |
+
"learning_rate": 8.675e-06,
|
| 326 |
+
"loss": 5.9192,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 6.65,
|
| 331 |
+
"learning_rate": 8.65e-06,
|
| 332 |
+
"loss": 5.8799,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.77,
|
| 337 |
+
"learning_rate": 8.625000000000001e-06,
|
| 338 |
+
"loss": 5.8367,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.89,
|
| 343 |
+
"learning_rate": 8.6e-06,
|
| 344 |
+
"loss": 5.7919,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 7.02,
|
| 349 |
+
"learning_rate": 8.575000000000002e-06,
|
| 350 |
+
"loss": 5.7692,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 7.14,
|
| 355 |
+
"learning_rate": 8.550000000000001e-06,
|
| 356 |
+
"loss": 5.7151,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 7.26,
|
| 361 |
+
"learning_rate": 8.525e-06,
|
| 362 |
+
"loss": 5.6839,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 7.38,
|
| 367 |
+
"learning_rate": 8.5e-06,
|
| 368 |
+
"loss": 5.6504,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 7.51,
|
| 373 |
+
"learning_rate": 8.475000000000001e-06,
|
| 374 |
+
"loss": 5.6222,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 7.63,
|
| 379 |
+
"learning_rate": 8.45e-06,
|
| 380 |
+
"loss": 5.5806,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 7.75,
|
| 385 |
+
"learning_rate": 8.425000000000001e-06,
|
| 386 |
+
"loss": 5.5354,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.88,
|
| 391 |
+
"learning_rate": 8.400000000000001e-06,
|
| 392 |
+
"loss": 5.4984,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 8.0,
|
| 397 |
+
"learning_rate": 8.375e-06,
|
| 398 |
+
"loss": 5.4788,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 8.12,
|
| 403 |
+
"learning_rate": 8.35e-06,
|
| 404 |
+
"loss": 5.4293,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 8.25,
|
| 409 |
+
"learning_rate": 8.325e-06,
|
| 410 |
+
"loss": 5.4256,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 8.37,
|
| 415 |
+
"learning_rate": 8.3e-06,
|
| 416 |
+
"loss": 5.3499,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 8.49,
|
| 421 |
+
"learning_rate": 8.275000000000001e-06,
|
| 422 |
+
"loss": 5.3758,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 8.62,
|
| 427 |
+
"learning_rate": 8.25e-06,
|
| 428 |
+
"loss": 5.3384,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 8.74,
|
| 433 |
+
"learning_rate": 8.225e-06,
|
| 434 |
+
"loss": 5.3189,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 8.86,
|
| 439 |
+
"learning_rate": 8.2e-06,
|
| 440 |
+
"loss": 5.2841,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 8.98,
|
| 445 |
+
"learning_rate": 8.175e-06,
|
| 446 |
+
"loss": 5.2387,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 9.11,
|
| 451 |
+
"learning_rate": 8.15e-06,
|
| 452 |
+
"loss": 5.229,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 9.23,
|
| 457 |
+
"learning_rate": 8.125000000000001e-06,
|
| 458 |
+
"loss": 5.1936,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 9.35,
|
| 463 |
+
"learning_rate": 8.1e-06,
|
| 464 |
+
"loss": 5.1758,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 9.48,
|
| 469 |
+
"learning_rate": 8.075000000000001e-06,
|
| 470 |
+
"loss": 5.1688,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 9.6,
|
| 475 |
+
"learning_rate": 8.050000000000001e-06,
|
| 476 |
+
"loss": 5.1475,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 9.72,
|
| 481 |
+
"learning_rate": 8.025e-06,
|
| 482 |
+
"loss": 5.1028,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 9.85,
|
| 487 |
+
"learning_rate": 8.000000000000001e-06,
|
| 488 |
+
"loss": 5.0747,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.97,
|
| 493 |
+
"learning_rate": 7.975e-06,
|
| 494 |
+
"loss": 5.0689,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 10.09,
|
| 499 |
+
"learning_rate": 7.950000000000002e-06,
|
| 500 |
+
"loss": 5.045,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 10.22,
|
| 505 |
+
"learning_rate": 7.925000000000001e-06,
|
| 506 |
+
"loss": 5.0361,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 10.34,
|
| 511 |
+
"learning_rate": 7.9e-06,
|
| 512 |
+
"loss": 5.0273,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 10.46,
|
| 517 |
+
"learning_rate": 7.875e-06,
|
| 518 |
+
"loss": 4.983,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 10.58,
|
| 523 |
+
"learning_rate": 7.850000000000001e-06,
|
| 524 |
+
"loss": 4.9598,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 10.71,
|
| 529 |
+
"learning_rate": 7.825e-06,
|
| 530 |
+
"loss": 4.9646,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 10.83,
|
| 535 |
+
"learning_rate": 7.800000000000002e-06,
|
| 536 |
+
"loss": 4.9083,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 10.95,
|
| 541 |
+
"learning_rate": 7.775000000000001e-06,
|
| 542 |
+
"loss": 4.906,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 11.08,
|
| 547 |
+
"learning_rate": 7.75e-06,
|
| 548 |
+
"loss": 4.9062,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 11.2,
|
| 553 |
+
"learning_rate": 7.725e-06,
|
| 554 |
+
"loss": 4.8742,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 11.32,
|
| 559 |
+
"learning_rate": 7.7e-06,
|
| 560 |
+
"loss": 4.8425,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 11.45,
|
| 565 |
+
"learning_rate": 7.675e-06,
|
| 566 |
+
"loss": 4.8332,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 11.57,
|
| 571 |
+
"learning_rate": 7.650000000000001e-06,
|
| 572 |
+
"loss": 4.8399,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 11.69,
|
| 577 |
+
"learning_rate": 7.625e-06,
|
| 578 |
+
"loss": 4.8392,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 11.82,
|
| 583 |
+
"learning_rate": 7.600000000000001e-06,
|
| 584 |
+
"loss": 4.8131,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 11.94,
|
| 589 |
+
"learning_rate": 7.575e-06,
|
| 590 |
+
"loss": 4.7643,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 12.06,
|
| 595 |
+
"learning_rate": 7.5500000000000006e-06,
|
| 596 |
+
"loss": 4.7836,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 12.18,
|
| 601 |
+
"learning_rate": 7.525e-06,
|
| 602 |
+
"loss": 4.7206,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 12.31,
|
| 607 |
+
"learning_rate": 7.500000000000001e-06,
|
| 608 |
+
"loss": 4.7083,
|
| 609 |
+
"step": 100
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 12.43,
|
| 613 |
+
"learning_rate": 7.475000000000001e-06,
|
| 614 |
+
"loss": 4.7409,
|
| 615 |
+
"step": 101
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 12.55,
|
| 619 |
+
"learning_rate": 7.450000000000001e-06,
|
| 620 |
+
"loss": 4.7328,
|
| 621 |
+
"step": 102
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 12.68,
|
| 625 |
+
"learning_rate": 7.425000000000001e-06,
|
| 626 |
+
"loss": 4.7065,
|
| 627 |
+
"step": 103
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 12.8,
|
| 631 |
+
"learning_rate": 7.4e-06,
|
| 632 |
+
"loss": 4.7223,
|
| 633 |
+
"step": 104
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 12.92,
|
| 637 |
+
"learning_rate": 7.375000000000001e-06,
|
| 638 |
+
"loss": 4.6878,
|
| 639 |
+
"step": 105
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 13.05,
|
| 643 |
+
"learning_rate": 7.350000000000001e-06,
|
| 644 |
+
"loss": 4.6661,
|
| 645 |
+
"step": 106
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 13.17,
|
| 649 |
+
"learning_rate": 7.325000000000001e-06,
|
| 650 |
+
"loss": 4.6621,
|
| 651 |
+
"step": 107
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 13.29,
|
| 655 |
+
"learning_rate": 7.3e-06,
|
| 656 |
+
"loss": 4.6306,
|
| 657 |
+
"step": 108
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 13.42,
|
| 661 |
+
"learning_rate": 7.275000000000001e-06,
|
| 662 |
+
"loss": 4.6375,
|
| 663 |
+
"step": 109
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 13.54,
|
| 667 |
+
"learning_rate": 7.25e-06,
|
| 668 |
+
"loss": 4.6358,
|
| 669 |
+
"step": 110
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 13.66,
|
| 673 |
+
"learning_rate": 7.225000000000001e-06,
|
| 674 |
+
"loss": 4.5903,
|
| 675 |
+
"step": 111
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 13.78,
|
| 679 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 680 |
+
"loss": 4.6222,
|
| 681 |
+
"step": 112
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 13.91,
|
| 685 |
+
"learning_rate": 7.175000000000001e-06,
|
| 686 |
+
"loss": 4.5943,
|
| 687 |
+
"step": 113
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 14.03,
|
| 691 |
+
"learning_rate": 7.15e-06,
|
| 692 |
+
"loss": 4.594,
|
| 693 |
+
"step": 114
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 14.15,
|
| 697 |
+
"learning_rate": 7.125e-06,
|
| 698 |
+
"loss": 4.5893,
|
| 699 |
+
"step": 115
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 14.28,
|
| 703 |
+
"learning_rate": 7.100000000000001e-06,
|
| 704 |
+
"loss": 4.5807,
|
| 705 |
+
"step": 116
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 14.4,
|
| 709 |
+
"learning_rate": 7.075000000000001e-06,
|
| 710 |
+
"loss": 4.5864,
|
| 711 |
+
"step": 117
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 14.52,
|
| 715 |
+
"learning_rate": 7.05e-06,
|
| 716 |
+
"loss": 4.5499,
|
| 717 |
+
"step": 118
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 14.65,
|
| 721 |
+
"learning_rate": 7.0250000000000005e-06,
|
| 722 |
+
"loss": 4.5211,
|
| 723 |
+
"step": 119
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 14.77,
|
| 727 |
+
"learning_rate": 7e-06,
|
| 728 |
+
"loss": 4.5026,
|
| 729 |
+
"step": 120
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 14.89,
|
| 733 |
+
"learning_rate": 6.975000000000001e-06,
|
| 734 |
+
"loss": 4.5158,
|
| 735 |
+
"step": 121
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 15.02,
|
| 739 |
+
"learning_rate": 6.95e-06,
|
| 740 |
+
"loss": 4.5286,
|
| 741 |
+
"step": 122
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 15.14,
|
| 745 |
+
"learning_rate": 6.925000000000001e-06,
|
| 746 |
+
"loss": 4.5167,
|
| 747 |
+
"step": 123
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 15.26,
|
| 751 |
+
"learning_rate": 6.9e-06,
|
| 752 |
+
"loss": 4.5259,
|
| 753 |
+
"step": 124
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 15.38,
|
| 757 |
+
"learning_rate": 6.875e-06,
|
| 758 |
+
"loss": 4.4632,
|
| 759 |
+
"step": 125
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 15.51,
|
| 763 |
+
"learning_rate": 6.850000000000001e-06,
|
| 764 |
+
"loss": 4.4855,
|
| 765 |
+
"step": 126
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 15.63,
|
| 769 |
+
"learning_rate": 6.825000000000001e-06,
|
| 770 |
+
"loss": 4.4902,
|
| 771 |
+
"step": 127
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 15.75,
|
| 775 |
+
"learning_rate": 6.800000000000001e-06,
|
| 776 |
+
"loss": 4.4322,
|
| 777 |
+
"step": 128
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 15.88,
|
| 781 |
+
"learning_rate": 6.775e-06,
|
| 782 |
+
"loss": 4.4617,
|
| 783 |
+
"step": 129
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 16.0,
|
| 787 |
+
"learning_rate": 6.750000000000001e-06,
|
| 788 |
+
"loss": 4.4418,
|
| 789 |
+
"step": 130
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 16.12,
|
| 793 |
+
"learning_rate": 6.725000000000001e-06,
|
| 794 |
+
"loss": 4.4433,
|
| 795 |
+
"step": 131
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 16.25,
|
| 799 |
+
"learning_rate": 6.700000000000001e-06,
|
| 800 |
+
"loss": 4.4117,
|
| 801 |
+
"step": 132
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 16.37,
|
| 805 |
+
"learning_rate": 6.6750000000000005e-06,
|
| 806 |
+
"loss": 4.433,
|
| 807 |
+
"step": 133
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 16.49,
|
| 811 |
+
"learning_rate": 6.650000000000001e-06,
|
| 812 |
+
"loss": 4.4145,
|
| 813 |
+
"step": 134
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 16.62,
|
| 817 |
+
"learning_rate": 6.625e-06,
|
| 818 |
+
"loss": 4.4216,
|
| 819 |
+
"step": 135
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 16.74,
|
| 823 |
+
"learning_rate": 6.600000000000001e-06,
|
| 824 |
+
"loss": 4.4239,
|
| 825 |
+
"step": 136
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 16.86,
|
| 829 |
+
"learning_rate": 6.5750000000000006e-06,
|
| 830 |
+
"loss": 4.4149,
|
| 831 |
+
"step": 137
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 16.98,
|
| 835 |
+
"learning_rate": 6.550000000000001e-06,
|
| 836 |
+
"loss": 4.4059,
|
| 837 |
+
"step": 138
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 17.11,
|
| 841 |
+
"learning_rate": 6.525e-06,
|
| 842 |
+
"loss": 4.3509,
|
| 843 |
+
"step": 139
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 17.23,
|
| 847 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 848 |
+
"loss": 4.4067,
|
| 849 |
+
"step": 140
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 17.35,
|
| 853 |
+
"learning_rate": 6.475e-06,
|
| 854 |
+
"loss": 4.3879,
|
| 855 |
+
"step": 141
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 17.48,
|
| 859 |
+
"learning_rate": 6.450000000000001e-06,
|
| 860 |
+
"loss": 4.3748,
|
| 861 |
+
"step": 142
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 17.6,
|
| 865 |
+
"learning_rate": 6.425e-06,
|
| 866 |
+
"loss": 4.3709,
|
| 867 |
+
"step": 143
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 17.72,
|
| 871 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 872 |
+
"loss": 4.3753,
|
| 873 |
+
"step": 144
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 17.85,
|
| 877 |
+
"learning_rate": 6.375e-06,
|
| 878 |
+
"loss": 4.3353,
|
| 879 |
+
"step": 145
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 17.97,
|
| 883 |
+
"learning_rate": 6.35e-06,
|
| 884 |
+
"loss": 4.3601,
|
| 885 |
+
"step": 146
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 18.09,
|
| 889 |
+
"learning_rate": 6.3250000000000004e-06,
|
| 890 |
+
"loss": 4.3194,
|
| 891 |
+
"step": 147
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 18.22,
|
| 895 |
+
"learning_rate": 6.300000000000001e-06,
|
| 896 |
+
"loss": 4.3581,
|
| 897 |
+
"step": 148
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 18.34,
|
| 901 |
+
"learning_rate": 6.275e-06,
|
| 902 |
+
"loss": 4.3432,
|
| 903 |
+
"step": 149
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 18.46,
|
| 907 |
+
"learning_rate": 6.25e-06,
|
| 908 |
+
"loss": 4.3122,
|
| 909 |
+
"step": 150
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 18.58,
|
| 913 |
+
"learning_rate": 6.225000000000001e-06,
|
| 914 |
+
"loss": 4.326,
|
| 915 |
+
"step": 151
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 18.71,
|
| 919 |
+
"learning_rate": 6.200000000000001e-06,
|
| 920 |
+
"loss": 4.3265,
|
| 921 |
+
"step": 152
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 18.83,
|
| 925 |
+
"learning_rate": 6.175000000000001e-06,
|
| 926 |
+
"loss": 4.3063,
|
| 927 |
+
"step": 153
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 18.95,
|
| 931 |
+
"learning_rate": 6.15e-06,
|
| 932 |
+
"loss": 4.2965,
|
| 933 |
+
"step": 154
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 19.08,
|
| 937 |
+
"learning_rate": 6.125000000000001e-06,
|
| 938 |
+
"loss": 4.277,
|
| 939 |
+
"step": 155
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 19.2,
|
| 943 |
+
"learning_rate": 6.1e-06,
|
| 944 |
+
"loss": 4.3039,
|
| 945 |
+
"step": 156
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 19.32,
|
| 949 |
+
"learning_rate": 6.075000000000001e-06,
|
| 950 |
+
"loss": 4.2957,
|
| 951 |
+
"step": 157
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 19.45,
|
| 955 |
+
"learning_rate": 6.0500000000000005e-06,
|
| 956 |
+
"loss": 4.2921,
|
| 957 |
+
"step": 158
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 19.57,
|
| 961 |
+
"learning_rate": 6.025000000000001e-06,
|
| 962 |
+
"loss": 4.2891,
|
| 963 |
+
"step": 159
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 19.69,
|
| 967 |
+
"learning_rate": 6e-06,
|
| 968 |
+
"loss": 4.276,
|
| 969 |
+
"step": 160
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 19.82,
|
| 973 |
+
"learning_rate": 5.975e-06,
|
| 974 |
+
"loss": 4.2549,
|
| 975 |
+
"step": 161
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 19.94,
|
| 979 |
+
"learning_rate": 5.950000000000001e-06,
|
| 980 |
+
"loss": 4.2798,
|
| 981 |
+
"step": 162
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 20.06,
|
| 985 |
+
"learning_rate": 5.925000000000001e-06,
|
| 986 |
+
"loss": 4.27,
|
| 987 |
+
"step": 163
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 20.18,
|
| 991 |
+
"learning_rate": 5.9e-06,
|
| 992 |
+
"loss": 4.238,
|
| 993 |
+
"step": 164
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 20.31,
|
| 997 |
+
"learning_rate": 5.8750000000000005e-06,
|
| 998 |
+
"loss": 4.2609,
|
| 999 |
+
"step": 165
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 20.43,
|
| 1003 |
+
"learning_rate": 5.85e-06,
|
| 1004 |
+
"loss": 4.2478,
|
| 1005 |
+
"step": 166
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 20.55,
|
| 1009 |
+
"learning_rate": 5.825000000000001e-06,
|
| 1010 |
+
"loss": 4.2445,
|
| 1011 |
+
"step": 167
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 20.68,
|
| 1015 |
+
"learning_rate": 5.8e-06,
|
| 1016 |
+
"loss": 4.2355,
|
| 1017 |
+
"step": 168
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 20.8,
|
| 1021 |
+
"learning_rate": 5.775000000000001e-06,
|
| 1022 |
+
"loss": 4.2569,
|
| 1023 |
+
"step": 169
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 20.92,
|
| 1027 |
+
"learning_rate": 5.75e-06,
|
| 1028 |
+
"loss": 4.25,
|
| 1029 |
+
"step": 170
|
| 1030 |
+
}
|
| 1031 |
+
],
|
| 1032 |
+
"logging_steps": 1,
|
| 1033 |
+
"max_steps": 400,
|
| 1034 |
+
"num_train_epochs": 50,
|
| 1035 |
+
"save_steps": 500,
|
| 1036 |
+
"total_flos": 1.1333763251306496e+16,
|
| 1037 |
+
"trial_name": null,
|
| 1038 |
+
"trial_params": null
|
| 1039 |
+
}
|
checkpoint-170/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60568d4dd0f6572abe8a02d69c2d1f0c25627e4687509d3e2f27d2eb514e2587
|
| 3 |
+
size 4600
|