Upload poetry ambi model
Browse files- checkpoint-2635/config.json +40 -0
- checkpoint-2635/model.safetensors +3 -0
- checkpoint-2635/optimizer.pt +3 -0
- checkpoint-2635/rng_state.pth +3 -0
- checkpoint-2635/scheduler.pt +3 -0
- checkpoint-2635/special_tokens_map.json +7 -0
- checkpoint-2635/tokenizer_config.json +58 -0
- checkpoint-2635/trainer_state.json +1875 -0
- checkpoint-2635/training_args.bin +3 -0
- checkpoint-2635/vocab.txt +0 -0
- config.json +40 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer_config.json +58 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
checkpoint-2635/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.3,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.3,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"LABEL_0": 0,
|
| 22 |
+
"LABEL_1": 1,
|
| 23 |
+
"LABEL_2": 2,
|
| 24 |
+
"LABEL_3": 3,
|
| 25 |
+
"LABEL_4": 4
|
| 26 |
+
},
|
| 27 |
+
"layer_norm_eps": 1e-12,
|
| 28 |
+
"max_position_embeddings": 512,
|
| 29 |
+
"model_type": "bert",
|
| 30 |
+
"num_attention_heads": 12,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"pad_token_id": 0,
|
| 33 |
+
"position_embedding_type": "absolute",
|
| 34 |
+
"problem_type": "single_label_classification",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.51.3",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 30522
|
| 40 |
+
}
|
checkpoint-2635/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd15cb5458e22263ee49fd81f1c4ceeb1ba82be843a220b2a47052a9ea4795fc
|
| 3 |
+
size 437967876
|
checkpoint-2635/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ab0c3d803b6a5c837a062add025e8f0a77a57aef0d0fe895675fac46d1bf8ae
|
| 3 |
+
size 876051194
|
checkpoint-2635/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ff14bf5205ed3616a875f3c22863cba5890da2f1dcf6e035a2b45a3b56fd4d5
|
| 3 |
+
size 13990
|
checkpoint-2635/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a306b77f4e3c46945e1f4bbf385d31f7a1532022fa6a36b8a828a162407bae30
|
| 3 |
+
size 1064
|
checkpoint-2635/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoint-2635/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
checkpoint-2635/trainer_state.json
ADDED
|
@@ -0,0 +1,1875 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 5.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2635,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.018975332068311195,
|
| 14 |
+
"grad_norm": 8.835118293762207,
|
| 15 |
+
"learning_rate": 1.0227272727272727e-06,
|
| 16 |
+
"loss": 1.7016,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03795066413662239,
|
| 21 |
+
"grad_norm": 8.580988883972168,
|
| 22 |
+
"learning_rate": 2.1590909090909092e-06,
|
| 23 |
+
"loss": 1.662,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.056925996204933584,
|
| 28 |
+
"grad_norm": 20.42087745666504,
|
| 29 |
+
"learning_rate": 3.2954545454545456e-06,
|
| 30 |
+
"loss": 1.6002,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07590132827324478,
|
| 35 |
+
"grad_norm": 10.94491195678711,
|
| 36 |
+
"learning_rate": 4.4318181818181824e-06,
|
| 37 |
+
"loss": 1.5688,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09487666034155598,
|
| 42 |
+
"grad_norm": 9.096759796142578,
|
| 43 |
+
"learning_rate": 5.568181818181818e-06,
|
| 44 |
+
"loss": 1.4954,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11385199240986717,
|
| 49 |
+
"grad_norm": 7.80708646774292,
|
| 50 |
+
"learning_rate": 6.704545454545454e-06,
|
| 51 |
+
"loss": 1.3318,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.13282732447817835,
|
| 56 |
+
"grad_norm": 6.090744972229004,
|
| 57 |
+
"learning_rate": 7.840909090909091e-06,
|
| 58 |
+
"loss": 1.1492,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.15180265654648956,
|
| 63 |
+
"grad_norm": 6.388769626617432,
|
| 64 |
+
"learning_rate": 8.977272727272727e-06,
|
| 65 |
+
"loss": 1.1073,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.17077798861480076,
|
| 70 |
+
"grad_norm": 5.182093143463135,
|
| 71 |
+
"learning_rate": 1.0113636363636363e-05,
|
| 72 |
+
"loss": 1.0731,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.18975332068311196,
|
| 77 |
+
"grad_norm": 6.343074798583984,
|
| 78 |
+
"learning_rate": 1.125e-05,
|
| 79 |
+
"loss": 1.058,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.20872865275142316,
|
| 84 |
+
"grad_norm": 5.188294410705566,
|
| 85 |
+
"learning_rate": 1.2386363636363637e-05,
|
| 86 |
+
"loss": 1.1108,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22770398481973433,
|
| 91 |
+
"grad_norm": 6.771484375,
|
| 92 |
+
"learning_rate": 1.3522727272727273e-05,
|
| 93 |
+
"loss": 1.1263,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.24667931688804554,
|
| 98 |
+
"grad_norm": 4.864952087402344,
|
| 99 |
+
"learning_rate": 1.465909090909091e-05,
|
| 100 |
+
"loss": 1.0434,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2656546489563567,
|
| 105 |
+
"grad_norm": 9.101506233215332,
|
| 106 |
+
"learning_rate": 1.5795454545454546e-05,
|
| 107 |
+
"loss": 1.0442,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.2846299810246679,
|
| 112 |
+
"grad_norm": 4.442895412445068,
|
| 113 |
+
"learning_rate": 1.6931818181818182e-05,
|
| 114 |
+
"loss": 1.1183,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.3036053130929791,
|
| 119 |
+
"grad_norm": 5.434806823730469,
|
| 120 |
+
"learning_rate": 1.8068181818181818e-05,
|
| 121 |
+
"loss": 1.0668,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3225806451612903,
|
| 126 |
+
"grad_norm": 3.6434082984924316,
|
| 127 |
+
"learning_rate": 1.9204545454545454e-05,
|
| 128 |
+
"loss": 0.9302,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3415559772296015,
|
| 133 |
+
"grad_norm": 5.795045852661133,
|
| 134 |
+
"learning_rate": 2.034090909090909e-05,
|
| 135 |
+
"loss": 1.0237,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3605313092979127,
|
| 140 |
+
"grad_norm": 3.951204538345337,
|
| 141 |
+
"learning_rate": 2.147727272727273e-05,
|
| 142 |
+
"loss": 0.9454,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3795066413662239,
|
| 147 |
+
"grad_norm": 4.916248798370361,
|
| 148 |
+
"learning_rate": 2.2613636363636365e-05,
|
| 149 |
+
"loss": 1.0498,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3984819734345351,
|
| 154 |
+
"grad_norm": 4.604029178619385,
|
| 155 |
+
"learning_rate": 2.3749999999999998e-05,
|
| 156 |
+
"loss": 0.9659,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4174573055028463,
|
| 161 |
+
"grad_norm": 3.4446909427642822,
|
| 162 |
+
"learning_rate": 2.4886363636363637e-05,
|
| 163 |
+
"loss": 0.9983,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.4364326375711575,
|
| 168 |
+
"grad_norm": 5.167362689971924,
|
| 169 |
+
"learning_rate": 2.6022727272727273e-05,
|
| 170 |
+
"loss": 1.1548,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.45540796963946867,
|
| 175 |
+
"grad_norm": 5.358937740325928,
|
| 176 |
+
"learning_rate": 2.715909090909091e-05,
|
| 177 |
+
"loss": 1.1427,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.47438330170777987,
|
| 182 |
+
"grad_norm": 4.279989242553711,
|
| 183 |
+
"learning_rate": 2.829545454545455e-05,
|
| 184 |
+
"loss": 1.1648,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.49335863377609107,
|
| 189 |
+
"grad_norm": 2.82234787940979,
|
| 190 |
+
"learning_rate": 2.9431818181818184e-05,
|
| 191 |
+
"loss": 1.0194,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.5123339658444023,
|
| 196 |
+
"grad_norm": 4.222620487213135,
|
| 197 |
+
"learning_rate": 2.9999670817816963e-05,
|
| 198 |
+
"loss": 1.105,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5313092979127134,
|
| 203 |
+
"grad_norm": 3.1942555904388428,
|
| 204 |
+
"learning_rate": 2.9997037447040725e-05,
|
| 205 |
+
"loss": 1.0327,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5502846299810247,
|
| 210 |
+
"grad_norm": 3.380519390106201,
|
| 211 |
+
"learning_rate": 2.9991771167807846e-05,
|
| 212 |
+
"loss": 1.0007,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5692599620493358,
|
| 217 |
+
"grad_norm": 3.5942091941833496,
|
| 218 |
+
"learning_rate": 2.9983872904676337e-05,
|
| 219 |
+
"loss": 1.1095,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5882352941176471,
|
| 224 |
+
"grad_norm": 4.9680256843566895,
|
| 225 |
+
"learning_rate": 2.997334404428031e-05,
|
| 226 |
+
"loss": 1.1684,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.6072106261859582,
|
| 231 |
+
"grad_norm": 7.381392002105713,
|
| 232 |
+
"learning_rate": 2.9960186435086557e-05,
|
| 233 |
+
"loss": 1.0933,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6261859582542695,
|
| 238 |
+
"grad_norm": 7.81553316116333,
|
| 239 |
+
"learning_rate": 2.9944402387069997e-05,
|
| 240 |
+
"loss": 1.0842,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6451612903225806,
|
| 245 |
+
"grad_norm": 2.633695363998413,
|
| 246 |
+
"learning_rate": 2.9925994671308168e-05,
|
| 247 |
+
"loss": 0.9835,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6641366223908919,
|
| 252 |
+
"grad_norm": 4.772572040557861,
|
| 253 |
+
"learning_rate": 2.9904966519494708e-05,
|
| 254 |
+
"loss": 1.0588,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.683111954459203,
|
| 259 |
+
"grad_norm": 3.716177463531494,
|
| 260 |
+
"learning_rate": 2.9881321623371996e-05,
|
| 261 |
+
"loss": 0.9534,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.7020872865275142,
|
| 266 |
+
"grad_norm": 3.3066067695617676,
|
| 267 |
+
"learning_rate": 2.9855064134083025e-05,
|
| 268 |
+
"loss": 0.9158,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.7210626185958254,
|
| 273 |
+
"grad_norm": 3.2305080890655518,
|
| 274 |
+
"learning_rate": 2.9826198661442628e-05,
|
| 275 |
+
"loss": 1.1518,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7400379506641366,
|
| 280 |
+
"grad_norm": 2.692566394805908,
|
| 281 |
+
"learning_rate": 2.9794730273128152e-05,
|
| 282 |
+
"loss": 0.9052,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7590132827324478,
|
| 287 |
+
"grad_norm": 4.057819843292236,
|
| 288 |
+
"learning_rate": 2.976066449378979e-05,
|
| 289 |
+
"loss": 0.9749,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.777988614800759,
|
| 294 |
+
"grad_norm": 3.3365478515625,
|
| 295 |
+
"learning_rate": 2.9724007304080635e-05,
|
| 296 |
+
"loss": 0.9746,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7969639468690702,
|
| 301 |
+
"grad_norm": 5.0923662185668945,
|
| 302 |
+
"learning_rate": 2.968476513960673e-05,
|
| 303 |
+
"loss": 0.9807,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.8159392789373814,
|
| 308 |
+
"grad_norm": 4.480515480041504,
|
| 309 |
+
"learning_rate": 2.9642944889797212e-05,
|
| 310 |
+
"loss": 1.0031,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8349146110056926,
|
| 315 |
+
"grad_norm": 3.221564769744873,
|
| 316 |
+
"learning_rate": 2.9598553896694777e-05,
|
| 317 |
+
"loss": 0.973,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8538899430740038,
|
| 322 |
+
"grad_norm": 5.767019748687744,
|
| 323 |
+
"learning_rate": 2.9551599953666725e-05,
|
| 324 |
+
"loss": 1.0251,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.872865275142315,
|
| 329 |
+
"grad_norm": 3.255776882171631,
|
| 330 |
+
"learning_rate": 2.950209130403672e-05,
|
| 331 |
+
"loss": 1.0173,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8918406072106262,
|
| 336 |
+
"grad_norm": 3.978546380996704,
|
| 337 |
+
"learning_rate": 2.945003663963758e-05,
|
| 338 |
+
"loss": 1.0973,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.9108159392789373,
|
| 343 |
+
"grad_norm": 3.4027981758117676,
|
| 344 |
+
"learning_rate": 2.9395445099285326e-05,
|
| 345 |
+
"loss": 0.9597,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.9297912713472486,
|
| 350 |
+
"grad_norm": 2.9666314125061035,
|
| 351 |
+
"learning_rate": 2.933832626717476e-05,
|
| 352 |
+
"loss": 0.9807,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9487666034155597,
|
| 357 |
+
"grad_norm": 6.159778118133545,
|
| 358 |
+
"learning_rate": 2.927869017119682e-05,
|
| 359 |
+
"loss": 1.1379,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.967741935483871,
|
| 364 |
+
"grad_norm": 3.3900465965270996,
|
| 365 |
+
"learning_rate": 2.921654728117811e-05,
|
| 366 |
+
"loss": 1.1136,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9867172675521821,
|
| 371 |
+
"grad_norm": 3.379819393157959,
|
| 372 |
+
"learning_rate": 2.9151908507042745e-05,
|
| 373 |
+
"loss": 1.165,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.0056925996204933,
|
| 378 |
+
"grad_norm": 2.686734437942505,
|
| 379 |
+
"learning_rate": 2.908478519689702e-05,
|
| 380 |
+
"loss": 0.9771,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.0246679316888045,
|
| 385 |
+
"grad_norm": 3.449796199798584,
|
| 386 |
+
"learning_rate": 2.901518913503711e-05,
|
| 387 |
+
"loss": 1.0079,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0436432637571158,
|
| 392 |
+
"grad_norm": 5.5144219398498535,
|
| 393 |
+
"learning_rate": 2.8943132539880174e-05,
|
| 394 |
+
"loss": 1.0533,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.0626185958254268,
|
| 399 |
+
"grad_norm": 3.1652843952178955,
|
| 400 |
+
"learning_rate": 2.8868628061819304e-05,
|
| 401 |
+
"loss": 1.0162,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.081593927893738,
|
| 406 |
+
"grad_norm": 3.358975887298584,
|
| 407 |
+
"learning_rate": 2.8791688781002553e-05,
|
| 408 |
+
"loss": 1.136,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.1005692599620494,
|
| 413 |
+
"grad_norm": 7.9716339111328125,
|
| 414 |
+
"learning_rate": 2.8712328205036603e-05,
|
| 415 |
+
"loss": 1.0258,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.1195445920303606,
|
| 420 |
+
"grad_norm": 2.9799373149871826,
|
| 421 |
+
"learning_rate": 2.8630560266615317e-05,
|
| 422 |
+
"loss": 0.9065,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.1385199240986716,
|
| 427 |
+
"grad_norm": 3.22654128074646,
|
| 428 |
+
"learning_rate": 2.854639932107369e-05,
|
| 429 |
+
"loss": 0.9845,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.157495256166983,
|
| 434 |
+
"grad_norm": 3.918231248855591,
|
| 435 |
+
"learning_rate": 2.8459860143867603e-05,
|
| 436 |
+
"loss": 1.1023,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.1764705882352942,
|
| 441 |
+
"grad_norm": 2.771279811859131,
|
| 442 |
+
"learning_rate": 2.8370957927979836e-05,
|
| 443 |
+
"loss": 1.0366,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.1954459203036052,
|
| 448 |
+
"grad_norm": 4.862953186035156,
|
| 449 |
+
"learning_rate": 2.827970828125272e-05,
|
| 450 |
+
"loss": 1.0149,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.2144212523719164,
|
| 455 |
+
"grad_norm": 5.020352363586426,
|
| 456 |
+
"learning_rate": 2.818612722364803e-05,
|
| 457 |
+
"loss": 1.0339,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.2333965844402277,
|
| 462 |
+
"grad_norm": 2.8869540691375732,
|
| 463 |
+
"learning_rate": 2.809023118443446e-05,
|
| 464 |
+
"loss": 0.9317,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.252371916508539,
|
| 469 |
+
"grad_norm": 3.7850170135498047,
|
| 470 |
+
"learning_rate": 2.7992036999303295e-05,
|
| 471 |
+
"loss": 1.0286,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.2713472485768502,
|
| 476 |
+
"grad_norm": 6.024351596832275,
|
| 477 |
+
"learning_rate": 2.7891561907412694e-05,
|
| 478 |
+
"loss": 1.132,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.2903225806451613,
|
| 483 |
+
"grad_norm": 4.305800437927246,
|
| 484 |
+
"learning_rate": 2.7788823548361144e-05,
|
| 485 |
+
"loss": 1.0266,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.3092979127134725,
|
| 490 |
+
"grad_norm": 3.822570562362671,
|
| 491 |
+
"learning_rate": 2.768383995909063e-05,
|
| 492 |
+
"loss": 1.0096,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.3282732447817835,
|
| 497 |
+
"grad_norm": 2.7487058639526367,
|
| 498 |
+
"learning_rate": 2.757662957072003e-05,
|
| 499 |
+
"loss": 0.9892,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.3472485768500948,
|
| 504 |
+
"grad_norm": 5.630655288696289,
|
| 505 |
+
"learning_rate": 2.7467211205309318e-05,
|
| 506 |
+
"loss": 0.9933,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.366223908918406,
|
| 511 |
+
"grad_norm": 3.3794660568237305,
|
| 512 |
+
"learning_rate": 2.7355604072555108e-05,
|
| 513 |
+
"loss": 1.0919,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.3851992409867173,
|
| 518 |
+
"grad_norm": 11.434050559997559,
|
| 519 |
+
"learning_rate": 2.7241827766418194e-05,
|
| 520 |
+
"loss": 1.0739,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.4041745730550286,
|
| 525 |
+
"grad_norm": 3.514951705932617,
|
| 526 |
+
"learning_rate": 2.712590226168357e-05,
|
| 527 |
+
"loss": 0.9835,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.4231499051233396,
|
| 532 |
+
"grad_norm": 6.354710102081299,
|
| 533 |
+
"learning_rate": 2.700784791045363e-05,
|
| 534 |
+
"loss": 0.9905,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.4421252371916509,
|
| 539 |
+
"grad_norm": 5.1164422035217285,
|
| 540 |
+
"learning_rate": 2.6887685438575116e-05,
|
| 541 |
+
"loss": 1.1429,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.4611005692599621,
|
| 546 |
+
"grad_norm": 3.0104331970214844,
|
| 547 |
+
"learning_rate": 2.676543594200045e-05,
|
| 548 |
+
"loss": 0.9537,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.4800759013282732,
|
| 553 |
+
"grad_norm": 7.312252044677734,
|
| 554 |
+
"learning_rate": 2.6641120883084068e-05,
|
| 555 |
+
"loss": 0.9569,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.4990512333965844,
|
| 560 |
+
"grad_norm": 3.7817318439483643,
|
| 561 |
+
"learning_rate": 2.6514762086814485e-05,
|
| 562 |
+
"loss": 1.07,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.5180265654648957,
|
| 567 |
+
"grad_norm": 4.462532997131348,
|
| 568 |
+
"learning_rate": 2.638638173698263e-05,
|
| 569 |
+
"loss": 1.0997,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.537001897533207,
|
| 574 |
+
"grad_norm": 4.784881591796875,
|
| 575 |
+
"learning_rate": 2.6256002372287226e-05,
|
| 576 |
+
"loss": 1.1417,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.5559772296015182,
|
| 581 |
+
"grad_norm": 5.281526565551758,
|
| 582 |
+
"learning_rate": 2.6123646882377844e-05,
|
| 583 |
+
"loss": 1.0294,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.5749525616698292,
|
| 588 |
+
"grad_norm": 5.754761695861816,
|
| 589 |
+
"learning_rate": 2.5989338503836376e-05,
|
| 590 |
+
"loss": 1.0487,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.5939278937381403,
|
| 595 |
+
"grad_norm": 4.882507801055908,
|
| 596 |
+
"learning_rate": 2.5853100816097547e-05,
|
| 597 |
+
"loss": 0.9086,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.6129032258064515,
|
| 602 |
+
"grad_norm": 5.188510894775391,
|
| 603 |
+
"learning_rate": 2.571495773730929e-05,
|
| 604 |
+
"loss": 1.0185,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.6318785578747628,
|
| 609 |
+
"grad_norm": 4.023245811462402,
|
| 610 |
+
"learning_rate": 2.5574933520133624e-05,
|
| 611 |
+
"loss": 0.9156,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.650853889943074,
|
| 616 |
+
"grad_norm": 4.175886154174805,
|
| 617 |
+
"learning_rate": 2.5433052747488782e-05,
|
| 618 |
+
"loss": 0.9871,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.6698292220113853,
|
| 623 |
+
"grad_norm": 6.542490005493164,
|
| 624 |
+
"learning_rate": 2.528934032823344e-05,
|
| 625 |
+
"loss": 1.0506,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.6888045540796965,
|
| 630 |
+
"grad_norm": 8.377986907958984,
|
| 631 |
+
"learning_rate": 2.5143821492793623e-05,
|
| 632 |
+
"loss": 1.0692,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.7077798861480076,
|
| 637 |
+
"grad_norm": 4.547092437744141,
|
| 638 |
+
"learning_rate": 2.4996521788733226e-05,
|
| 639 |
+
"loss": 1.0796,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.7267552182163188,
|
| 644 |
+
"grad_norm": 4.059111595153809,
|
| 645 |
+
"learning_rate": 2.4847467076268837e-05,
|
| 646 |
+
"loss": 0.983,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.7457305502846299,
|
| 651 |
+
"grad_norm": 3.2969706058502197,
|
| 652 |
+
"learning_rate": 2.469668352372965e-05,
|
| 653 |
+
"loss": 1.082,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.7647058823529411,
|
| 658 |
+
"grad_norm": 3.242330551147461,
|
| 659 |
+
"learning_rate": 2.4544197602963327e-05,
|
| 660 |
+
"loss": 1.0638,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.7836812144212524,
|
| 665 |
+
"grad_norm": 3.480397939682007,
|
| 666 |
+
"learning_rate": 2.4390036084688536e-05,
|
| 667 |
+
"loss": 0.9762,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.8026565464895636,
|
| 672 |
+
"grad_norm": 4.5360493659973145,
|
| 673 |
+
"learning_rate": 2.4234226033795022e-05,
|
| 674 |
+
"loss": 1.0095,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.821631878557875,
|
| 679 |
+
"grad_norm": 3.487361431121826,
|
| 680 |
+
"learning_rate": 2.4076794804592073e-05,
|
| 681 |
+
"loss": 1.1234,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.840607210626186,
|
| 686 |
+
"grad_norm": 3.7385318279266357,
|
| 687 |
+
"learning_rate": 2.391777003600612e-05,
|
| 688 |
+
"loss": 1.0511,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.8595825426944972,
|
| 693 |
+
"grad_norm": 5.168221473693848,
|
| 694 |
+
"learning_rate": 2.3757179646728408e-05,
|
| 695 |
+
"loss": 1.0301,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.8785578747628082,
|
| 700 |
+
"grad_norm": 3.248091459274292,
|
| 701 |
+
"learning_rate": 2.359505183031352e-05,
|
| 702 |
+
"loss": 0.9788,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.8975332068311195,
|
| 707 |
+
"grad_norm": 5.787475109100342,
|
| 708 |
+
"learning_rate": 2.3431415050229696e-05,
|
| 709 |
+
"loss": 1.066,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.9165085388994307,
|
| 714 |
+
"grad_norm": 5.45881986618042,
|
| 715 |
+
"learning_rate": 2.32662980348617e-05,
|
| 716 |
+
"loss": 0.9832,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.935483870967742,
|
| 721 |
+
"grad_norm": 6.038186073303223,
|
| 722 |
+
"learning_rate": 2.309972977246721e-05,
|
| 723 |
+
"loss": 0.9953,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.9544592030360532,
|
| 728 |
+
"grad_norm": 6.9032697677612305,
|
| 729 |
+
"learning_rate": 2.2931739506087613e-05,
|
| 730 |
+
"loss": 0.9966,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.9734345351043643,
|
| 735 |
+
"grad_norm": 5.0695648193359375,
|
| 736 |
+
"learning_rate": 2.2762356728414027e-05,
|
| 737 |
+
"loss": 1.0886,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.9924098671726755,
|
| 742 |
+
"grad_norm": 3.990579843521118,
|
| 743 |
+
"learning_rate": 2.2591611176609486e-05,
|
| 744 |
+
"loss": 1.1455,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.0113851992409866,
|
| 749 |
+
"grad_norm": 4.9777421951293945,
|
| 750 |
+
"learning_rate": 2.241953282708826e-05,
|
| 751 |
+
"loss": 1.0095,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.030360531309298,
|
| 756 |
+
"grad_norm": 5.180793762207031,
|
| 757 |
+
"learning_rate": 2.2246151890253125e-05,
|
| 758 |
+
"loss": 0.9768,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.049335863377609,
|
| 763 |
+
"grad_norm": 5.023456573486328,
|
| 764 |
+
"learning_rate": 2.207149880519155e-05,
|
| 765 |
+
"loss": 1.0083,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.0683111954459203,
|
| 770 |
+
"grad_norm": 3.7236404418945312,
|
| 771 |
+
"learning_rate": 2.18956042343318e-05,
|
| 772 |
+
"loss": 1.039,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.0872865275142316,
|
| 777 |
+
"grad_norm": 4.609447956085205,
|
| 778 |
+
"learning_rate": 2.171849905805972e-05,
|
| 779 |
+
"loss": 1.0211,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.106261859582543,
|
| 784 |
+
"grad_norm": 3.467719078063965,
|
| 785 |
+
"learning_rate": 2.1540214369297365e-05,
|
| 786 |
+
"loss": 1.0291,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.1252371916508537,
|
| 791 |
+
"grad_norm": 6.9950852394104,
|
| 792 |
+
"learning_rate": 2.136078146804424e-05,
|
| 793 |
+
"loss": 1.0414,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.144212523719165,
|
| 798 |
+
"grad_norm": 5.30154275894165,
|
| 799 |
+
"learning_rate": 2.118023185588223e-05,
|
| 800 |
+
"loss": 1.0929,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.163187855787476,
|
| 805 |
+
"grad_norm": 7.387163162231445,
|
| 806 |
+
"learning_rate": 2.0998597230445116e-05,
|
| 807 |
+
"loss": 0.983,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.1821631878557874,
|
| 812 |
+
"grad_norm": 5.720730781555176,
|
| 813 |
+
"learning_rate": 2.0815909479853654e-05,
|
| 814 |
+
"loss": 1.0384,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.2011385199240987,
|
| 819 |
+
"grad_norm": 5.67633581161499,
|
| 820 |
+
"learning_rate": 2.0632200677117283e-05,
|
| 821 |
+
"loss": 1.0319,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 2.22011385199241,
|
| 826 |
+
"grad_norm": 7.9937005043029785,
|
| 827 |
+
"learning_rate": 2.0447503074503295e-05,
|
| 828 |
+
"loss": 0.9644,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.239089184060721,
|
| 833 |
+
"grad_norm": 7.219088077545166,
|
| 834 |
+
"learning_rate": 2.0261849097874566e-05,
|
| 835 |
+
"loss": 1.0004,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.258064516129032,
|
| 840 |
+
"grad_norm": 4.610164642333984,
|
| 841 |
+
"learning_rate": 2.0075271340996818e-05,
|
| 842 |
+
"loss": 1.0091,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.2770398481973433,
|
| 847 |
+
"grad_norm": 4.246598243713379,
|
| 848 |
+
"learning_rate": 1.988780255981639e-05,
|
| 849 |
+
"loss": 1.0254,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 2.2960151802656545,
|
| 854 |
+
"grad_norm": 5.702746391296387,
|
| 855 |
+
"learning_rate": 1.969947566670955e-05,
|
| 856 |
+
"loss": 1.0655,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.314990512333966,
|
| 861 |
+
"grad_norm": 4.958011627197266,
|
| 862 |
+
"learning_rate": 1.951032372470433e-05,
|
| 863 |
+
"loss": 0.9928,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.333965844402277,
|
| 868 |
+
"grad_norm": 7.796271324157715,
|
| 869 |
+
"learning_rate": 1.932037994167591e-05,
|
| 870 |
+
"loss": 1.0571,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 2.3529411764705883,
|
| 875 |
+
"grad_norm": 5.767906188964844,
|
| 876 |
+
"learning_rate": 1.9129677664516603e-05,
|
| 877 |
+
"loss": 0.9137,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.3719165085388996,
|
| 882 |
+
"grad_norm": 5.057350158691406,
|
| 883 |
+
"learning_rate": 1.8938250373281377e-05,
|
| 884 |
+
"loss": 0.9793,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.3908918406072104,
|
| 889 |
+
"grad_norm": 4.325188159942627,
|
| 890 |
+
"learning_rate": 1.8746131675310066e-05,
|
| 891 |
+
"loss": 0.9912,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.4098671726755216,
|
| 896 |
+
"grad_norm": 5.487137794494629,
|
| 897 |
+
"learning_rate": 1.8553355299327187e-05,
|
| 898 |
+
"loss": 0.9622,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.428842504743833,
|
| 903 |
+
"grad_norm": 5.010653495788574,
|
| 904 |
+
"learning_rate": 1.8359955089520456e-05,
|
| 905 |
+
"loss": 0.9217,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.447817836812144,
|
| 910 |
+
"grad_norm": 4.641795635223389,
|
| 911 |
+
"learning_rate": 1.8165964999599038e-05,
|
| 912 |
+
"loss": 0.901,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.4667931688804554,
|
| 917 |
+
"grad_norm": 5.894201755523682,
|
| 918 |
+
"learning_rate": 1.797141908683259e-05,
|
| 919 |
+
"loss": 0.918,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.4857685009487667,
|
| 924 |
+
"grad_norm": 5.431739330291748,
|
| 925 |
+
"learning_rate": 1.7776351506072073e-05,
|
| 926 |
+
"loss": 0.904,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.504743833017078,
|
| 931 |
+
"grad_norm": 5.612817287445068,
|
| 932 |
+
"learning_rate": 1.758079650375348e-05,
|
| 933 |
+
"loss": 0.9759,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.5237191650853887,
|
| 938 |
+
"grad_norm": 7.929321765899658,
|
| 939 |
+
"learning_rate": 1.7384788411885465e-05,
|
| 940 |
+
"loss": 1.0011,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.5426944971537004,
|
| 945 |
+
"grad_norm": 4.3894782066345215,
|
| 946 |
+
"learning_rate": 1.718836164202193e-05,
|
| 947 |
+
"loss": 0.9004,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.5616698292220113,
|
| 952 |
+
"grad_norm": 4.695666790008545,
|
| 953 |
+
"learning_rate": 1.699155067922071e-05,
|
| 954 |
+
"loss": 0.9371,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.5806451612903225,
|
| 959 |
+
"grad_norm": 8.929224014282227,
|
| 960 |
+
"learning_rate": 1.6794390075989253e-05,
|
| 961 |
+
"loss": 1.1086,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 2.5996204933586338,
|
| 966 |
+
"grad_norm": 5.96087646484375,
|
| 967 |
+
"learning_rate": 1.659691444621857e-05,
|
| 968 |
+
"loss": 0.9386,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.618595825426945,
|
| 973 |
+
"grad_norm": 7.858741283416748,
|
| 974 |
+
"learning_rate": 1.6399158459106327e-05,
|
| 975 |
+
"loss": 1.0329,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.6375711574952563,
|
| 980 |
+
"grad_norm": 5.565644264221191,
|
| 981 |
+
"learning_rate": 1.620115683307025e-05,
|
| 982 |
+
"loss": 1.0101,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.656546489563567,
|
| 987 |
+
"grad_norm": 5.135579586029053,
|
| 988 |
+
"learning_rate": 1.6002944329652906e-05,
|
| 989 |
+
"loss": 1.0569,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 2.675521821631879,
|
| 994 |
+
"grad_norm": 4.882253646850586,
|
| 995 |
+
"learning_rate": 1.5804555747418906e-05,
|
| 996 |
+
"loss": 1.0134,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 2.6944971537001896,
|
| 1001 |
+
"grad_norm": 6.51032018661499,
|
| 1002 |
+
"learning_rate": 1.560602591584558e-05,
|
| 1003 |
+
"loss": 1.0513,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.713472485768501,
|
| 1008 |
+
"grad_norm": 4.702990531921387,
|
| 1009 |
+
"learning_rate": 1.5407389689208264e-05,
|
| 1010 |
+
"loss": 0.9764,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 2.732447817836812,
|
| 1015 |
+
"grad_norm": 4.451667308807373,
|
| 1016 |
+
"learning_rate": 1.5208681940461227e-05,
|
| 1017 |
+
"loss": 0.8766,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 2.7514231499051234,
|
| 1022 |
+
"grad_norm": 4.740606784820557,
|
| 1023 |
+
"learning_rate": 1.500993755511528e-05,
|
| 1024 |
+
"loss": 1.009,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 2.7703984819734346,
|
| 1029 |
+
"grad_norm": 4.939847469329834,
|
| 1030 |
+
"learning_rate": 1.4811191425113227e-05,
|
| 1031 |
+
"loss": 1.0194,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 2.789373814041746,
|
| 1036 |
+
"grad_norm": 4.593587875366211,
|
| 1037 |
+
"learning_rate": 1.4612478442704176e-05,
|
| 1038 |
+
"loss": 0.9864,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 2.808349146110057,
|
| 1043 |
+
"grad_norm": 4.617082595825195,
|
| 1044 |
+
"learning_rate": 1.4413833494317775e-05,
|
| 1045 |
+
"loss": 0.8882,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.827324478178368,
|
| 1050 |
+
"grad_norm": 4.513877868652344,
|
| 1051 |
+
"learning_rate": 1.4215291454439486e-05,
|
| 1052 |
+
"loss": 0.9707,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 2.846299810246679,
|
| 1057 |
+
"grad_norm": 6.98907470703125,
|
| 1058 |
+
"learning_rate": 1.4016887179487966e-05,
|
| 1059 |
+
"loss": 0.9479,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 2.8652751423149905,
|
| 1064 |
+
"grad_norm": 4.818065166473389,
|
| 1065 |
+
"learning_rate": 1.3818655501695587e-05,
|
| 1066 |
+
"loss": 1.069,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.8842504743833017,
|
| 1071 |
+
"grad_norm": 5.875453948974609,
|
| 1072 |
+
"learning_rate": 1.362063122299325e-05,
|
| 1073 |
+
"loss": 0.9822,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.903225806451613,
|
| 1078 |
+
"grad_norm": 5.649266719818115,
|
| 1079 |
+
"learning_rate": 1.3422849108900467e-05,
|
| 1080 |
+
"loss": 0.9977,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.9222011385199242,
|
| 1085 |
+
"grad_norm": 5.480551242828369,
|
| 1086 |
+
"learning_rate": 1.3225343882421872e-05,
|
| 1087 |
+
"loss": 1.1059,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.9411764705882355,
|
| 1092 |
+
"grad_norm": 4.593949317932129,
|
| 1093 |
+
"learning_rate": 1.3028150217951176e-05,
|
| 1094 |
+
"loss": 1.0366,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 2.9601518026565463,
|
| 1099 |
+
"grad_norm": 7.213682651519775,
|
| 1100 |
+
"learning_rate": 1.2831302735183682e-05,
|
| 1101 |
+
"loss": 0.9752,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 2.9791271347248576,
|
| 1106 |
+
"grad_norm": 7.539572715759277,
|
| 1107 |
+
"learning_rate": 1.2634835993038372e-05,
|
| 1108 |
+
"loss": 0.883,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.998102466793169,
|
| 1113 |
+
"grad_norm": 5.167015552520752,
|
| 1114 |
+
"learning_rate": 1.2438784483590667e-05,
|
| 1115 |
+
"loss": 0.9209,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 3.01707779886148,
|
| 1120 |
+
"grad_norm": 5.3906168937683105,
|
| 1121 |
+
"learning_rate": 1.224318262601696e-05,
|
| 1122 |
+
"loss": 0.9377,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 3.0360531309297913,
|
| 1127 |
+
"grad_norm": 7.1217780113220215,
|
| 1128 |
+
"learning_rate": 1.2048064760551894e-05,
|
| 1129 |
+
"loss": 0.9382,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 3.0550284629981026,
|
| 1134 |
+
"grad_norm": 5.268466472625732,
|
| 1135 |
+
"learning_rate": 1.1853465142459541e-05,
|
| 1136 |
+
"loss": 0.949,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 3.074003795066414,
|
| 1141 |
+
"grad_norm": 6.088480472564697,
|
| 1142 |
+
"learning_rate": 1.1659417936019478e-05,
|
| 1143 |
+
"loss": 0.9653,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 3.0929791271347247,
|
| 1148 |
+
"grad_norm": 5.27643346786499,
|
| 1149 |
+
"learning_rate": 1.1465957208528835e-05,
|
| 1150 |
+
"loss": 1.0644,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 3.111954459203036,
|
| 1155 |
+
"grad_norm": 5.9762420654296875,
|
| 1156 |
+
"learning_rate": 1.1273116924321383e-05,
|
| 1157 |
+
"loss": 0.886,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 3.130929791271347,
|
| 1162 |
+
"grad_norm": 7.467761993408203,
|
| 1163 |
+
"learning_rate": 1.1080930938804693e-05,
|
| 1164 |
+
"loss": 0.8474,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 3.1499051233396584,
|
| 1169 |
+
"grad_norm": 5.725830078125,
|
| 1170 |
+
"learning_rate": 1.0889432992516406e-05,
|
| 1171 |
+
"loss": 1.0272,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 3.1688804554079697,
|
| 1176 |
+
"grad_norm": 8.115575790405273,
|
| 1177 |
+
"learning_rate": 1.069865670520069e-05,
|
| 1178 |
+
"loss": 0.9451,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 3.187855787476281,
|
| 1183 |
+
"grad_norm": 7.985052585601807,
|
| 1184 |
+
"learning_rate": 1.0508635569905883e-05,
|
| 1185 |
+
"loss": 1.1718,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 3.206831119544592,
|
| 1190 |
+
"grad_norm": 5.023871421813965,
|
| 1191 |
+
"learning_rate": 1.03194029471044e-05,
|
| 1192 |
+
"loss": 0.9923,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 3.225806451612903,
|
| 1197 |
+
"grad_norm": 8.017499923706055,
|
| 1198 |
+
"learning_rate": 1.0130992058835904e-05,
|
| 1199 |
+
"loss": 0.9651,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 3.2447817836812143,
|
| 1204 |
+
"grad_norm": 6.202470779418945,
|
| 1205 |
+
"learning_rate": 9.943435982874775e-06,
|
| 1206 |
+
"loss": 0.8275,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 3.2637571157495255,
|
| 1211 |
+
"grad_norm": 7.41610050201416,
|
| 1212 |
+
"learning_rate": 9.756767646922915e-06,
|
| 1213 |
+
"loss": 0.8881,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.282732447817837,
|
| 1218 |
+
"grad_norm": 7.807908535003662,
|
| 1219 |
+
"learning_rate": 9.5710198228289e-06,
|
| 1220 |
+
"loss": 0.9826,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 3.301707779886148,
|
| 1225 |
+
"grad_norm": 8.9044771194458,
|
| 1226 |
+
"learning_rate": 9.386225120834497e-06,
|
| 1227 |
+
"loss": 1.0274,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 3.3206831119544593,
|
| 1232 |
+
"grad_norm": 6.678797245025635,
|
| 1233 |
+
"learning_rate": 9.202415983849529e-06,
|
| 1234 |
+
"loss": 0.959,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 3.3396584440227706,
|
| 1239 |
+
"grad_norm": 7.571164608001709,
|
| 1240 |
+
"learning_rate": 9.019624681756165e-06,
|
| 1241 |
+
"loss": 0.8161,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 3.3586337760910814,
|
| 1246 |
+
"grad_norm": 6.871535778045654,
|
| 1247 |
+
"learning_rate": 8.837883305743528e-06,
|
| 1248 |
+
"loss": 1.0435,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 3.3776091081593926,
|
| 1253 |
+
"grad_norm": 7.78122091293335,
|
| 1254 |
+
"learning_rate": 8.657223762673728e-06,
|
| 1255 |
+
"loss": 0.9105,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 3.396584440227704,
|
| 1260 |
+
"grad_norm": 7.536012649536133,
|
| 1261 |
+
"learning_rate": 8.477677769480215e-06,
|
| 1262 |
+
"loss": 1.1449,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 3.415559772296015,
|
| 1267 |
+
"grad_norm": 8.20406723022461,
|
| 1268 |
+
"learning_rate": 8.299276847599504e-06,
|
| 1269 |
+
"loss": 0.9355,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 3.4345351043643264,
|
| 1274 |
+
"grad_norm": 10.926506996154785,
|
| 1275 |
+
"learning_rate": 8.122052317437223e-06,
|
| 1276 |
+
"loss": 0.9103,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 3.4535104364326377,
|
| 1281 |
+
"grad_norm": 7.018414497375488,
|
| 1282 |
+
"learning_rate": 7.946035292869411e-06,
|
| 1283 |
+
"loss": 0.923,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 3.472485768500949,
|
| 1288 |
+
"grad_norm": 6.66978645324707,
|
| 1289 |
+
"learning_rate": 7.771256675780134e-06,
|
| 1290 |
+
"loss": 0.9447,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 3.4914611005692597,
|
| 1295 |
+
"grad_norm": 7.309480667114258,
|
| 1296 |
+
"learning_rate": 7.597747150636283e-06,
|
| 1297 |
+
"loss": 0.8562,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 3.510436432637571,
|
| 1302 |
+
"grad_norm": 8.151505470275879,
|
| 1303 |
+
"learning_rate": 7.425537179100542e-06,
|
| 1304 |
+
"loss": 1.0463,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 3.5294117647058822,
|
| 1309 |
+
"grad_norm": 6.514859199523926,
|
| 1310 |
+
"learning_rate": 7.254656994683497e-06,
|
| 1311 |
+
"loss": 0.8109,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 3.5483870967741935,
|
| 1316 |
+
"grad_norm": 7.040705680847168,
|
| 1317 |
+
"learning_rate": 7.085136597435775e-06,
|
| 1318 |
+
"loss": 0.9602,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 3.5673624288425048,
|
| 1323 |
+
"grad_norm": 6.3895087242126465,
|
| 1324 |
+
"learning_rate": 6.917005748681172e-06,
|
| 1325 |
+
"loss": 0.9311,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 3.586337760910816,
|
| 1330 |
+
"grad_norm": 6.2356743812561035,
|
| 1331 |
+
"learning_rate": 6.750293965791702e-06,
|
| 1332 |
+
"loss": 0.9547,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 3.6053130929791273,
|
| 1337 |
+
"grad_norm": 8.55842399597168,
|
| 1338 |
+
"learning_rate": 6.585030517005491e-06,
|
| 1339 |
+
"loss": 0.7918,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 3.624288425047438,
|
| 1344 |
+
"grad_norm": 10.336483001708984,
|
| 1345 |
+
"learning_rate": 6.421244416288351e-06,
|
| 1346 |
+
"loss": 0.9755,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 3.64326375711575,
|
| 1351 |
+
"grad_norm": 6.095067024230957,
|
| 1352 |
+
"learning_rate": 6.258964418240062e-06,
|
| 1353 |
+
"loss": 1.0992,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 3.6622390891840606,
|
| 1358 |
+
"grad_norm": 9.044556617736816,
|
| 1359 |
+
"learning_rate": 6.098219013046149e-06,
|
| 1360 |
+
"loss": 0.9321,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 3.681214421252372,
|
| 1365 |
+
"grad_norm": 7.915792465209961,
|
| 1366 |
+
"learning_rate": 5.939036421476094e-06,
|
| 1367 |
+
"loss": 0.8914,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 3.700189753320683,
|
| 1372 |
+
"grad_norm": 7.116144180297852,
|
| 1373 |
+
"learning_rate": 5.781444589928843e-06,
|
| 1374 |
+
"loss": 0.8809,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 3.7191650853889944,
|
| 1379 |
+
"grad_norm": 7.079509258270264,
|
| 1380 |
+
"learning_rate": 5.625471185526483e-06,
|
| 1381 |
+
"loss": 0.9792,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 3.7381404174573056,
|
| 1386 |
+
"grad_norm": 7.109429359436035,
|
| 1387 |
+
"learning_rate": 5.471143591256947e-06,
|
| 1388 |
+
"loss": 0.9313,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 3.7571157495256164,
|
| 1393 |
+
"grad_norm": 7.368680477142334,
|
| 1394 |
+
"learning_rate": 5.3184889011666015e-06,
|
| 1395 |
+
"loss": 0.9071,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 3.776091081593928,
|
| 1400 |
+
"grad_norm": 7.896711826324463,
|
| 1401 |
+
"learning_rate": 5.167533915603589e-06,
|
| 1402 |
+
"loss": 0.8605,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 3.795066413662239,
|
| 1407 |
+
"grad_norm": 6.464845657348633,
|
| 1408 |
+
"learning_rate": 5.018305136512667e-06,
|
| 1409 |
+
"loss": 0.8312,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 3.81404174573055,
|
| 1414 |
+
"grad_norm": 9.47612190246582,
|
| 1415 |
+
"learning_rate": 4.870828762782513e-06,
|
| 1416 |
+
"loss": 0.9728,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 3.8330170777988615,
|
| 1421 |
+
"grad_norm": 5.801048278808594,
|
| 1422 |
+
"learning_rate": 4.7251306856461575e-06,
|
| 1423 |
+
"loss": 0.835,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 3.8519924098671727,
|
| 1428 |
+
"grad_norm": 8.155261039733887,
|
| 1429 |
+
"learning_rate": 4.581236484135521e-06,
|
| 1430 |
+
"loss": 0.9336,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 3.870967741935484,
|
| 1435 |
+
"grad_norm": 7.239008903503418,
|
| 1436 |
+
"learning_rate": 4.43917142059067e-06,
|
| 1437 |
+
"loss": 1.0022,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 3.889943074003795,
|
| 1442 |
+
"grad_norm": 9.633106231689453,
|
| 1443 |
+
"learning_rate": 4.298960436224725e-06,
|
| 1444 |
+
"loss": 0.9201,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 3.9089184060721065,
|
| 1449 |
+
"grad_norm": 10.326329231262207,
|
| 1450 |
+
"learning_rate": 4.160628146745148e-06,
|
| 1451 |
+
"loss": 0.9183,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 3.9278937381404173,
|
| 1456 |
+
"grad_norm": 6.792428016662598,
|
| 1457 |
+
"learning_rate": 4.0241988380321345e-06,
|
| 1458 |
+
"loss": 0.9018,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 3.9468690702087286,
|
| 1463 |
+
"grad_norm": 9.20457649230957,
|
| 1464 |
+
"learning_rate": 3.889696461874956e-06,
|
| 1465 |
+
"loss": 0.9354,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.96584440227704,
|
| 1470 |
+
"grad_norm": 14.224815368652344,
|
| 1471 |
+
"learning_rate": 3.757144631766939e-06,
|
| 1472 |
+
"loss": 1.1006,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 3.984819734345351,
|
| 1477 |
+
"grad_norm": 8.766386985778809,
|
| 1478 |
+
"learning_rate": 3.626566618759824e-06,
|
| 1479 |
+
"loss": 0.8878,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 4.003795066413662,
|
| 1484 |
+
"grad_norm": 6.551325798034668,
|
| 1485 |
+
"learning_rate": 3.497985347378265e-06,
|
| 1486 |
+
"loss": 0.8765,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 4.022770398481973,
|
| 1491 |
+
"grad_norm": 8.314547538757324,
|
| 1492 |
+
"learning_rate": 3.371423391595157e-06,
|
| 1493 |
+
"loss": 0.9016,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 4.041745730550285,
|
| 1498 |
+
"grad_norm": 8.383026123046875,
|
| 1499 |
+
"learning_rate": 3.246902970868496e-06,
|
| 1500 |
+
"loss": 0.904,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 4.060721062618596,
|
| 1505 |
+
"grad_norm": 7.667434215545654,
|
| 1506 |
+
"learning_rate": 3.124445946240499e-06,
|
| 1507 |
+
"loss": 0.8713,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 4.079696394686907,
|
| 1512 |
+
"grad_norm": 7.005184650421143,
|
| 1513 |
+
"learning_rate": 3.004073816499637e-06,
|
| 1514 |
+
"loss": 0.9774,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 4.098671726755218,
|
| 1519 |
+
"grad_norm": 7.62594747543335,
|
| 1520 |
+
"learning_rate": 2.885807714406261e-06,
|
| 1521 |
+
"loss": 1.0262,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 4.117647058823529,
|
| 1526 |
+
"grad_norm": 7.611791133880615,
|
| 1527 |
+
"learning_rate": 2.7696684029824937e-06,
|
| 1528 |
+
"loss": 0.8275,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 4.136622390891841,
|
| 1533 |
+
"grad_norm": 8.397326469421387,
|
| 1534 |
+
"learning_rate": 2.655676271867042e-06,
|
| 1535 |
+
"loss": 0.9084,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 4.1555977229601515,
|
| 1540 |
+
"grad_norm": 10.16572380065918,
|
| 1541 |
+
"learning_rate": 2.543851333735547e-06,
|
| 1542 |
+
"loss": 0.974,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 4.174573055028463,
|
| 1547 |
+
"grad_norm": 11.483360290527344,
|
| 1548 |
+
"learning_rate": 2.4342132207871215e-06,
|
| 1549 |
+
"loss": 0.9486,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 4.193548387096774,
|
| 1554 |
+
"grad_norm": 7.796977996826172,
|
| 1555 |
+
"learning_rate": 2.3267811812976856e-06,
|
| 1556 |
+
"loss": 0.9159,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 4.212523719165086,
|
| 1561 |
+
"grad_norm": 9.091662406921387,
|
| 1562 |
+
"learning_rate": 2.2215740762407e-06,
|
| 1563 |
+
"loss": 0.8267,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 4.2314990512333965,
|
| 1568 |
+
"grad_norm": 7.4472784996032715,
|
| 1569 |
+
"learning_rate": 2.118610375975896e-06,
|
| 1570 |
+
"loss": 0.9187,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 4.250474383301707,
|
| 1575 |
+
"grad_norm": 7.37217903137207,
|
| 1576 |
+
"learning_rate": 2.017908157006597e-06,
|
| 1577 |
+
"loss": 0.8676,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 4.269449715370019,
|
| 1582 |
+
"grad_norm": 8.160748481750488,
|
| 1583 |
+
"learning_rate": 1.9194850988061547e-06,
|
| 1584 |
+
"loss": 0.9814,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 4.28842504743833,
|
| 1589 |
+
"grad_norm": 7.1200337409973145,
|
| 1590 |
+
"learning_rate": 1.8233584807141156e-06,
|
| 1591 |
+
"loss": 0.8773,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 4.307400379506642,
|
| 1596 |
+
"grad_norm": 7.349186897277832,
|
| 1597 |
+
"learning_rate": 1.7295451789026285e-06,
|
| 1598 |
+
"loss": 0.8973,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 4.326375711574952,
|
| 1603 |
+
"grad_norm": 6.584765434265137,
|
| 1604 |
+
"learning_rate": 1.638061663413633e-06,
|
| 1605 |
+
"loss": 0.8918,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 4.345351043643264,
|
| 1610 |
+
"grad_norm": 9.584866523742676,
|
| 1611 |
+
"learning_rate": 1.5489239952673401e-06,
|
| 1612 |
+
"loss": 0.9519,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 4.364326375711575,
|
| 1617 |
+
"grad_norm": 9.459647178649902,
|
| 1618 |
+
"learning_rate": 1.4621478236425356e-06,
|
| 1619 |
+
"loss": 0.9792,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 4.383301707779886,
|
| 1624 |
+
"grad_norm": 9.50500774383545,
|
| 1625 |
+
"learning_rate": 1.377748383129181e-06,
|
| 1626 |
+
"loss": 0.9223,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 4.402277039848197,
|
| 1631 |
+
"grad_norm": 8.901126861572266,
|
| 1632 |
+
"learning_rate": 1.2957404910537902e-06,
|
| 1633 |
+
"loss": 0.8686,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 4.421252371916508,
|
| 1638 |
+
"grad_norm": 8.882048606872559,
|
| 1639 |
+
"learning_rate": 1.2161385448780877e-06,
|
| 1640 |
+
"loss": 1.0387,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 4.44022770398482,
|
| 1645 |
+
"grad_norm": 8.212772369384766,
|
| 1646 |
+
"learning_rate": 1.1389565196713485e-06,
|
| 1647 |
+
"loss": 0.8612,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 4.459203036053131,
|
| 1652 |
+
"grad_norm": 8.24146556854248,
|
| 1653 |
+
"learning_rate": 1.0642079656569114e-06,
|
| 1654 |
+
"loss": 0.9589,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 4.478178368121442,
|
| 1659 |
+
"grad_norm": 7.321491718292236,
|
| 1660 |
+
"learning_rate": 9.91906005833284e-07,
|
| 1661 |
+
"loss": 0.89,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 4.497153700189753,
|
| 1666 |
+
"grad_norm": 8.948949813842773,
|
| 1667 |
+
"learning_rate": 9.220633336702334e-07,
|
| 1668 |
+
"loss": 0.7829,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 4.516129032258064,
|
| 1673 |
+
"grad_norm": 6.95490026473999,
|
| 1674 |
+
"learning_rate": 8.546922108803029e-07,
|
| 1675 |
+
"loss": 0.85,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 4.535104364326376,
|
| 1680 |
+
"grad_norm": 8.235620498657227,
|
| 1681 |
+
"learning_rate": 7.898044652661185e-07,
|
| 1682 |
+
"loss": 0.9012,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 4.554079696394687,
|
| 1687 |
+
"grad_norm": 9.577425956726074,
|
| 1688 |
+
"learning_rate": 7.274114886438837e-07,
|
| 1689 |
+
"loss": 0.9101,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 4.573055028462998,
|
| 1694 |
+
"grad_norm": 7.038369655609131,
|
| 1695 |
+
"learning_rate": 6.675242348434018e-07,
|
| 1696 |
+
"loss": 0.9156,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 4.592030360531309,
|
| 1701 |
+
"grad_norm": 6.6995015144348145,
|
| 1702 |
+
"learning_rate": 6.101532177850106e-07,
|
| 1703 |
+
"loss": 0.8572,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 4.611005692599621,
|
| 1708 |
+
"grad_norm": 9.94770622253418,
|
| 1709 |
+
"learning_rate": 5.553085096337285e-07,
|
| 1710 |
+
"loss": 0.9834,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 4.629981024667932,
|
| 1715 |
+
"grad_norm": 7.077449798583984,
|
| 1716 |
+
"learning_rate": 5.029997390309748e-07,
|
| 1717 |
+
"loss": 0.7711,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 4.648956356736242,
|
| 1722 |
+
"grad_norm": 8.839421272277832,
|
| 1723 |
+
"learning_rate": 4.532360894041365e-07,
|
| 1724 |
+
"loss": 0.892,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 4.667931688804554,
|
| 1729 |
+
"grad_norm": 7.479890823364258,
|
| 1730 |
+
"learning_rate": 4.0602629735432273e-07,
|
| 1731 |
+
"loss": 1.0207,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 4.686907020872865,
|
| 1736 |
+
"grad_norm": 9.129741668701172,
|
| 1737 |
+
"learning_rate": 3.6137865112253533e-07,
|
| 1738 |
+
"loss": 0.9569,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 4.705882352941177,
|
| 1743 |
+
"grad_norm": 6.612678050994873,
|
| 1744 |
+
"learning_rate": 3.1930098913458273e-07,
|
| 1745 |
+
"loss": 0.7619,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 4.724857685009487,
|
| 1750 |
+
"grad_norm": 8.517851829528809,
|
| 1751 |
+
"learning_rate": 2.7980069862493905e-07,
|
| 1752 |
+
"loss": 0.92,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 4.743833017077799,
|
| 1757 |
+
"grad_norm": 7.410338401794434,
|
| 1758 |
+
"learning_rate": 2.4288471433983826e-07,
|
| 1759 |
+
"loss": 0.8749,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 4.76280834914611,
|
| 1764 |
+
"grad_norm": 9.296804428100586,
|
| 1765 |
+
"learning_rate": 2.0855951731978962e-07,
|
| 1766 |
+
"loss": 0.9728,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 4.781783681214421,
|
| 1771 |
+
"grad_norm": 6.3679986000061035,
|
| 1772 |
+
"learning_rate": 1.768311337617562e-07,
|
| 1773 |
+
"loss": 0.9575,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 4.8007590132827325,
|
| 1778 |
+
"grad_norm": 9.791215896606445,
|
| 1779 |
+
"learning_rate": 1.4770513396118124e-07,
|
| 1780 |
+
"loss": 0.8171,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 4.819734345351043,
|
| 1785 |
+
"grad_norm": 11.64243221282959,
|
| 1786 |
+
"learning_rate": 1.2118663133406204e-07,
|
| 1787 |
+
"loss": 0.9217,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 4.838709677419355,
|
| 1792 |
+
"grad_norm": 8.724605560302734,
|
| 1793 |
+
"learning_rate": 9.728028151922475e-08,
|
| 1794 |
+
"loss": 0.9398,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 4.857685009487666,
|
| 1799 |
+
"grad_norm": 6.529393196105957,
|
| 1800 |
+
"learning_rate": 7.599028156097643e-08,
|
| 1801 |
+
"loss": 0.9897,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 4.8766603415559775,
|
| 1806 |
+
"grad_norm": 10.065555572509766,
|
| 1807 |
+
"learning_rate": 5.732036917225292e-08,
|
| 1808 |
+
"loss": 0.7993,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 4.895635673624288,
|
| 1813 |
+
"grad_norm": 6.991254806518555,
|
| 1814 |
+
"learning_rate": 4.127382207843533e-08,
|
| 1815 |
+
"loss": 0.8928,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 4.914611005692599,
|
| 1820 |
+
"grad_norm": 8.686799049377441,
|
| 1821 |
+
"learning_rate": 2.785345744188872e-08,
|
| 1822 |
+
"loss": 0.9171,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 4.933586337760911,
|
| 1827 |
+
"grad_norm": 8.544384956359863,
|
| 1828 |
+
"learning_rate": 1.7061631367389363e-08,
|
| 1829 |
+
"loss": 0.9,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 4.952561669829222,
|
| 1834 |
+
"grad_norm": 11.076976776123047,
|
| 1835 |
+
"learning_rate": 8.900238488478984e-09,
|
| 1836 |
+
"loss": 1.0119,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 4.971537001897533,
|
| 1841 |
+
"grad_norm": 10.223211288452148,
|
| 1842 |
+
"learning_rate": 3.3707116348324907e-09,
|
| 1843 |
+
"loss": 0.9601,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 4.990512333965844,
|
| 1848 |
+
"grad_norm": 8.19120979309082,
|
| 1849 |
+
"learning_rate": 4.740215807158554e-10,
|
| 1850 |
+
"loss": 0.801,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
}
|
| 1853 |
+
],
|
| 1854 |
+
"logging_steps": 10,
|
| 1855 |
+
"max_steps": 2635,
|
| 1856 |
+
"num_input_tokens_seen": 0,
|
| 1857 |
+
"num_train_epochs": 5,
|
| 1858 |
+
"save_steps": 500,
|
| 1859 |
+
"stateful_callbacks": {
|
| 1860 |
+
"TrainerControl": {
|
| 1861 |
+
"args": {
|
| 1862 |
+
"should_epoch_stop": false,
|
| 1863 |
+
"should_evaluate": false,
|
| 1864 |
+
"should_log": false,
|
| 1865 |
+
"should_save": true,
|
| 1866 |
+
"should_training_stop": true
|
| 1867 |
+
},
|
| 1868 |
+
"attributes": {}
|
| 1869 |
+
}
|
| 1870 |
+
},
|
| 1871 |
+
"total_flos": 2771291835502080.0,
|
| 1872 |
+
"train_batch_size": 8,
|
| 1873 |
+
"trial_name": null,
|
| 1874 |
+
"trial_params": null
|
| 1875 |
+
}
|
checkpoint-2635/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3aa5021973dbef1c7dba1d6da264e20ca0835902b5a372d2e3b9df4dd6c0f6a7
|
| 3 |
+
size 5240
|
checkpoint-2635/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.3,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.3,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"LABEL_0": 0,
|
| 22 |
+
"LABEL_1": 1,
|
| 23 |
+
"LABEL_2": 2,
|
| 24 |
+
"LABEL_3": 3,
|
| 25 |
+
"LABEL_4": 4
|
| 26 |
+
},
|
| 27 |
+
"layer_norm_eps": 1e-12,
|
| 28 |
+
"max_position_embeddings": 512,
|
| 29 |
+
"model_type": "bert",
|
| 30 |
+
"num_attention_heads": 12,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"pad_token_id": 0,
|
| 33 |
+
"position_embedding_type": "absolute",
|
| 34 |
+
"problem_type": "single_label_classification",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.51.3",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 30522
|
| 40 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd15cb5458e22263ee49fd81f1c4ceeb1ba82be843a220b2a47052a9ea4795fc
|
| 3 |
+
size 437967876
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3aa5021973dbef1c7dba1d6da264e20ca0835902b5a372d2e3b9df4dd6c0f6a7
|
| 3 |
+
size 5240
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|