feat: add tokenizer
Browse filesThis view is limited to 50 files because it contains too many changes. 聽 See raw diff
- checkpoint-1000/config.json +27 -0
- checkpoint-1000/model.safetensors +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scaler.pt +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +37 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer_config.json +56 -0
- checkpoint-1000/trainer_state.json +363 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-1000/vocab.txt +0 -0
- checkpoint-1500/config.json +27 -0
- checkpoint-1500/model.safetensors +3 -0
- checkpoint-1500/optimizer.pt +3 -0
- checkpoint-1500/rng_state.pth +3 -0
- checkpoint-1500/scaler.pt +3 -0
- checkpoint-1500/scheduler.pt +3 -0
- checkpoint-1500/special_tokens_map.json +37 -0
- checkpoint-1500/tokenizer.json +0 -0
- checkpoint-1500/tokenizer_config.json +56 -0
- checkpoint-1500/trainer_state.json +523 -0
- checkpoint-1500/training_args.bin +3 -0
- checkpoint-1500/vocab.txt +0 -0
- checkpoint-2000/config.json +27 -0
- checkpoint-2000/model.safetensors +3 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scaler.pt +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +37 -0
- checkpoint-2000/tokenizer.json +0 -0
- checkpoint-2000/tokenizer_config.json +56 -0
- checkpoint-2000/trainer_state.json +683 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-2000/vocab.txt +0 -0
- checkpoint-2500/config.json +27 -0
- checkpoint-2500/model.safetensors +3 -0
- checkpoint-2500/optimizer.pt +3 -0
- checkpoint-2500/rng_state.pth +3 -0
- checkpoint-2500/scaler.pt +3 -0
- checkpoint-2500/scheduler.pt +3 -0
- checkpoint-2500/special_tokens_map.json +37 -0
- checkpoint-2500/tokenizer.json +0 -0
- checkpoint-2500/tokenizer_config.json +56 -0
- checkpoint-2500/trainer_state.json +843 -0
- checkpoint-2500/training_args.bin +3 -0
- checkpoint-2500/vocab.txt +0 -0
- checkpoint-3000/config.json +27 -0
- checkpoint-3000/model.safetensors +3 -0
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForQuestionAnswering"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"language": "english",
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"name": "Bert",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-1000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f6e5a45458a20c806ca603db7cba95ce6c5ba90e8163c7908bbf3ef1077fd2d
|
| 3 |
+
size 435596088
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e56b503363dd06ab1f3e37a59f7f01a8eff9100346afb12b2df580c75c0a9e14
|
| 3 |
+
size 871311930
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9eb0e8e8bc64936aefc31ee7af1a5d1912fdfb5b07d689677f79a6259ca2c757
|
| 3 |
+
size 14244
|
checkpoint-1000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38ca2316913aaacc2d40490a6e90fc16b3fba3b663945bbfe95df2edfb678f0a
|
| 3 |
+
size 988
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:473b1e2df25ac8c3e9e7d7a54e92cafd4a272f4381022c30a6f20583c6587137
|
| 3 |
+
size 1064
|
checkpoint-1000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-1000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,363 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 900,
|
| 3 |
+
"best_metric": 82.39127215176644,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.26490066225165565,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013245033112582781,
|
| 14 |
+
"grad_norm": 7.202547550201416,
|
| 15 |
+
"learning_rate": 9.998701986754968e-06,
|
| 16 |
+
"loss": 0.7094,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.026490066225165563,
|
| 21 |
+
"grad_norm": 9.662558555603027,
|
| 22 |
+
"learning_rate": 9.99737748344371e-06,
|
| 23 |
+
"loss": 0.68,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.026490066225165563,
|
| 28 |
+
"eval_HasAns_exact": 75.14792899408285,
|
| 29 |
+
"eval_HasAns_f1": 81.01957227969451,
|
| 30 |
+
"eval_HasAns_total": 15041,
|
| 31 |
+
"eval_best_exact": 75.14792899408285,
|
| 32 |
+
"eval_best_exact_thresh": 0.0,
|
| 33 |
+
"eval_best_f1": 81.01957227969451,
|
| 34 |
+
"eval_best_f1_thresh": 0.0,
|
| 35 |
+
"eval_exact": 75.14792899408285,
|
| 36 |
+
"eval_f1": 81.01957227969451,
|
| 37 |
+
"eval_loss": 0.5824803113937378,
|
| 38 |
+
"eval_runtime": 120.1578,
|
| 39 |
+
"eval_samples_per_second": 125.177,
|
| 40 |
+
"eval_steps_per_second": 7.831,
|
| 41 |
+
"eval_total": 15041,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.039735099337748346,
|
| 46 |
+
"grad_norm": 7.527211666107178,
|
| 47 |
+
"learning_rate": 9.996052980132451e-06,
|
| 48 |
+
"loss": 0.586,
|
| 49 |
+
"step": 150
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.052980132450331126,
|
| 53 |
+
"grad_norm": 5.419323921203613,
|
| 54 |
+
"learning_rate": 9.994728476821193e-06,
|
| 55 |
+
"loss": 0.6035,
|
| 56 |
+
"step": 200
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.052980132450331126,
|
| 60 |
+
"eval_HasAns_exact": 75.56013562927997,
|
| 61 |
+
"eval_HasAns_f1": 81.46153699536843,
|
| 62 |
+
"eval_HasAns_total": 15041,
|
| 63 |
+
"eval_best_exact": 75.56013562927997,
|
| 64 |
+
"eval_best_exact_thresh": 0.0,
|
| 65 |
+
"eval_best_f1": 81.46153699536843,
|
| 66 |
+
"eval_best_f1_thresh": 0.0,
|
| 67 |
+
"eval_exact": 75.56013562927997,
|
| 68 |
+
"eval_f1": 81.46153699536843,
|
| 69 |
+
"eval_loss": 0.5453085899353027,
|
| 70 |
+
"eval_runtime": 118.3719,
|
| 71 |
+
"eval_samples_per_second": 127.066,
|
| 72 |
+
"eval_steps_per_second": 7.95,
|
| 73 |
+
"eval_total": 15041,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.06622516556291391,
|
| 78 |
+
"grad_norm": 9.9099760055542,
|
| 79 |
+
"learning_rate": 9.993403973509935e-06,
|
| 80 |
+
"loss": 0.5616,
|
| 81 |
+
"step": 250
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.07947019867549669,
|
| 85 |
+
"grad_norm": 3.5184719562530518,
|
| 86 |
+
"learning_rate": 9.992079470198677e-06,
|
| 87 |
+
"loss": 0.5286,
|
| 88 |
+
"step": 300
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.07947019867549669,
|
| 92 |
+
"eval_HasAns_exact": 75.8526693703876,
|
| 93 |
+
"eval_HasAns_f1": 81.45745729278944,
|
| 94 |
+
"eval_HasAns_total": 15041,
|
| 95 |
+
"eval_best_exact": 75.8526693703876,
|
| 96 |
+
"eval_best_exact_thresh": 0.0,
|
| 97 |
+
"eval_best_f1": 81.45745729278944,
|
| 98 |
+
"eval_best_f1_thresh": 0.0,
|
| 99 |
+
"eval_exact": 75.8526693703876,
|
| 100 |
+
"eval_f1": 81.45745729278944,
|
| 101 |
+
"eval_loss": 0.5200360417366028,
|
| 102 |
+
"eval_runtime": 118.7441,
|
| 103 |
+
"eval_samples_per_second": 126.667,
|
| 104 |
+
"eval_steps_per_second": 7.925,
|
| 105 |
+
"eval_total": 15041,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.09271523178807947,
|
| 110 |
+
"grad_norm": 4.889588832855225,
|
| 111 |
+
"learning_rate": 9.990754966887418e-06,
|
| 112 |
+
"loss": 0.4943,
|
| 113 |
+
"step": 350
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.10596026490066225,
|
| 117 |
+
"grad_norm": 6.106881141662598,
|
| 118 |
+
"learning_rate": 9.98943046357616e-06,
|
| 119 |
+
"loss": 0.5957,
|
| 120 |
+
"step": 400
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.10596026490066225,
|
| 124 |
+
"eval_HasAns_exact": 76.39784588790639,
|
| 125 |
+
"eval_HasAns_f1": 81.91199737367272,
|
| 126 |
+
"eval_HasAns_total": 15041,
|
| 127 |
+
"eval_best_exact": 76.39784588790639,
|
| 128 |
+
"eval_best_exact_thresh": 0.0,
|
| 129 |
+
"eval_best_f1": 81.91199737367272,
|
| 130 |
+
"eval_best_f1_thresh": 0.0,
|
| 131 |
+
"eval_exact": 76.39784588790639,
|
| 132 |
+
"eval_f1": 81.91199737367272,
|
| 133 |
+
"eval_loss": 0.49782633781433105,
|
| 134 |
+
"eval_runtime": 118.2655,
|
| 135 |
+
"eval_samples_per_second": 127.18,
|
| 136 |
+
"eval_steps_per_second": 7.957,
|
| 137 |
+
"eval_total": 15041,
|
| 138 |
+
"step": 400
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.11920529801324503,
|
| 142 |
+
"grad_norm": 6.963775634765625,
|
| 143 |
+
"learning_rate": 9.988105960264902e-06,
|
| 144 |
+
"loss": 0.5675,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.13245033112582782,
|
| 149 |
+
"grad_norm": 7.559121608734131,
|
| 150 |
+
"learning_rate": 9.986781456953643e-06,
|
| 151 |
+
"loss": 0.5173,
|
| 152 |
+
"step": 500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.13245033112582782,
|
| 156 |
+
"eval_HasAns_exact": 75.94574828801277,
|
| 157 |
+
"eval_HasAns_f1": 81.43615773338746,
|
| 158 |
+
"eval_HasAns_total": 15041,
|
| 159 |
+
"eval_best_exact": 75.94574828801277,
|
| 160 |
+
"eval_best_exact_thresh": 0.0,
|
| 161 |
+
"eval_best_f1": 81.43615773338746,
|
| 162 |
+
"eval_best_f1_thresh": 0.0,
|
| 163 |
+
"eval_exact": 75.94574828801277,
|
| 164 |
+
"eval_f1": 81.43615773338746,
|
| 165 |
+
"eval_loss": 0.4918788969516754,
|
| 166 |
+
"eval_runtime": 118.4997,
|
| 167 |
+
"eval_samples_per_second": 126.929,
|
| 168 |
+
"eval_steps_per_second": 7.941,
|
| 169 |
+
"eval_total": 15041,
|
| 170 |
+
"step": 500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1456953642384106,
|
| 174 |
+
"grad_norm": 5.2273268699646,
|
| 175 |
+
"learning_rate": 9.985456953642385e-06,
|
| 176 |
+
"loss": 0.5059,
|
| 177 |
+
"step": 550
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.15894039735099338,
|
| 181 |
+
"grad_norm": 13.281010627746582,
|
| 182 |
+
"learning_rate": 9.984132450331127e-06,
|
| 183 |
+
"loss": 0.4847,
|
| 184 |
+
"step": 600
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.15894039735099338,
|
| 188 |
+
"eval_HasAns_exact": 76.42443986437073,
|
| 189 |
+
"eval_HasAns_f1": 81.98556495448281,
|
| 190 |
+
"eval_HasAns_total": 15041,
|
| 191 |
+
"eval_best_exact": 76.42443986437073,
|
| 192 |
+
"eval_best_exact_thresh": 0.0,
|
| 193 |
+
"eval_best_f1": 81.98556495448281,
|
| 194 |
+
"eval_best_f1_thresh": 0.0,
|
| 195 |
+
"eval_exact": 76.42443986437073,
|
| 196 |
+
"eval_f1": 81.98556495448281,
|
| 197 |
+
"eval_loss": 0.48446840047836304,
|
| 198 |
+
"eval_runtime": 118.3928,
|
| 199 |
+
"eval_samples_per_second": 127.043,
|
| 200 |
+
"eval_steps_per_second": 7.948,
|
| 201 |
+
"eval_total": 15041,
|
| 202 |
+
"step": 600
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.17218543046357615,
|
| 206 |
+
"grad_norm": 18.346189498901367,
|
| 207 |
+
"learning_rate": 9.982807947019869e-06,
|
| 208 |
+
"loss": 0.4857,
|
| 209 |
+
"step": 650
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.18543046357615894,
|
| 213 |
+
"grad_norm": 11.825461387634277,
|
| 214 |
+
"learning_rate": 9.98148344370861e-06,
|
| 215 |
+
"loss": 0.4764,
|
| 216 |
+
"step": 700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.18543046357615894,
|
| 220 |
+
"eval_HasAns_exact": 76.54411275846022,
|
| 221 |
+
"eval_HasAns_f1": 82.14175205695277,
|
| 222 |
+
"eval_HasAns_total": 15041,
|
| 223 |
+
"eval_best_exact": 76.54411275846022,
|
| 224 |
+
"eval_best_exact_thresh": 0.0,
|
| 225 |
+
"eval_best_f1": 82.14175205695277,
|
| 226 |
+
"eval_best_f1_thresh": 0.0,
|
| 227 |
+
"eval_exact": 76.54411275846022,
|
| 228 |
+
"eval_f1": 82.14175205695277,
|
| 229 |
+
"eval_loss": 0.4717336595058441,
|
| 230 |
+
"eval_runtime": 118.5022,
|
| 231 |
+
"eval_samples_per_second": 126.926,
|
| 232 |
+
"eval_steps_per_second": 7.941,
|
| 233 |
+
"eval_total": 15041,
|
| 234 |
+
"step": 700
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1986754966887417,
|
| 238 |
+
"grad_norm": 7.6593475341796875,
|
| 239 |
+
"learning_rate": 9.980158940397352e-06,
|
| 240 |
+
"loss": 0.486,
|
| 241 |
+
"step": 750
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2119205298013245,
|
| 245 |
+
"grad_norm": 6.091643333435059,
|
| 246 |
+
"learning_rate": 9.978834437086094e-06,
|
| 247 |
+
"loss": 0.4707,
|
| 248 |
+
"step": 800
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2119205298013245,
|
| 252 |
+
"eval_HasAns_exact": 76.86324047603217,
|
| 253 |
+
"eval_HasAns_f1": 82.2239211673303,
|
| 254 |
+
"eval_HasAns_total": 15041,
|
| 255 |
+
"eval_best_exact": 76.86324047603217,
|
| 256 |
+
"eval_best_exact_thresh": 0.0,
|
| 257 |
+
"eval_best_f1": 82.2239211673303,
|
| 258 |
+
"eval_best_f1_thresh": 0.0,
|
| 259 |
+
"eval_exact": 76.86324047603217,
|
| 260 |
+
"eval_f1": 82.2239211673303,
|
| 261 |
+
"eval_loss": 0.4601329565048218,
|
| 262 |
+
"eval_runtime": 119.1574,
|
| 263 |
+
"eval_samples_per_second": 126.228,
|
| 264 |
+
"eval_steps_per_second": 7.897,
|
| 265 |
+
"eval_total": 15041,
|
| 266 |
+
"step": 800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2251655629139073,
|
| 270 |
+
"grad_norm": 4.863158702850342,
|
| 271 |
+
"learning_rate": 9.977509933774835e-06,
|
| 272 |
+
"loss": 0.5042,
|
| 273 |
+
"step": 850
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.23841059602649006,
|
| 277 |
+
"grad_norm": 5.747822284698486,
|
| 278 |
+
"learning_rate": 9.976185430463577e-06,
|
| 279 |
+
"loss": 0.4725,
|
| 280 |
+
"step": 900
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.23841059602649006,
|
| 284 |
+
"eval_HasAns_exact": 76.98291337012166,
|
| 285 |
+
"eval_HasAns_f1": 82.39127215176644,
|
| 286 |
+
"eval_HasAns_total": 15041,
|
| 287 |
+
"eval_best_exact": 76.98291337012166,
|
| 288 |
+
"eval_best_exact_thresh": 0.0,
|
| 289 |
+
"eval_best_f1": 82.39127215176644,
|
| 290 |
+
"eval_best_f1_thresh": 0.0,
|
| 291 |
+
"eval_exact": 76.98291337012166,
|
| 292 |
+
"eval_f1": 82.39127215176644,
|
| 293 |
+
"eval_loss": 0.45768168568611145,
|
| 294 |
+
"eval_runtime": 118.7264,
|
| 295 |
+
"eval_samples_per_second": 126.686,
|
| 296 |
+
"eval_steps_per_second": 7.926,
|
| 297 |
+
"eval_total": 15041,
|
| 298 |
+
"step": 900
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.25165562913907286,
|
| 302 |
+
"grad_norm": 14.890397071838379,
|
| 303 |
+
"learning_rate": 9.974860927152319e-06,
|
| 304 |
+
"loss": 0.4476,
|
| 305 |
+
"step": 950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.26490066225165565,
|
| 309 |
+
"grad_norm": 8.837732315063477,
|
| 310 |
+
"learning_rate": 9.97353642384106e-06,
|
| 311 |
+
"loss": 0.4905,
|
| 312 |
+
"step": 1000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.26490066225165565,
|
| 316 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 317 |
+
"eval_HasAns_f1": 82.37799175072134,
|
| 318 |
+
"eval_HasAns_total": 15041,
|
| 319 |
+
"eval_best_exact": 77.10923475832723,
|
| 320 |
+
"eval_best_exact_thresh": 0.0,
|
| 321 |
+
"eval_best_f1": 82.37799175072134,
|
| 322 |
+
"eval_best_f1_thresh": 0.0,
|
| 323 |
+
"eval_exact": 77.10923475832723,
|
| 324 |
+
"eval_f1": 82.37799175072134,
|
| 325 |
+
"eval_loss": 0.44477367401123047,
|
| 326 |
+
"eval_runtime": 118.302,
|
| 327 |
+
"eval_samples_per_second": 127.141,
|
| 328 |
+
"eval_steps_per_second": 7.954,
|
| 329 |
+
"eval_total": 15041,
|
| 330 |
+
"step": 1000
|
| 331 |
+
}
|
| 332 |
+
],
|
| 333 |
+
"logging_steps": 50,
|
| 334 |
+
"max_steps": 377500,
|
| 335 |
+
"num_input_tokens_seen": 0,
|
| 336 |
+
"num_train_epochs": 100,
|
| 337 |
+
"save_steps": 500,
|
| 338 |
+
"stateful_callbacks": {
|
| 339 |
+
"EarlyStoppingCallback": {
|
| 340 |
+
"args": {
|
| 341 |
+
"early_stopping_patience": 10,
|
| 342 |
+
"early_stopping_threshold": 0.0
|
| 343 |
+
},
|
| 344 |
+
"attributes": {
|
| 345 |
+
"early_stopping_patience_counter": 1
|
| 346 |
+
}
|
| 347 |
+
},
|
| 348 |
+
"TrainerControl": {
|
| 349 |
+
"args": {
|
| 350 |
+
"should_epoch_stop": false,
|
| 351 |
+
"should_evaluate": false,
|
| 352 |
+
"should_log": false,
|
| 353 |
+
"should_save": true,
|
| 354 |
+
"should_training_stop": false
|
| 355 |
+
},
|
| 356 |
+
"attributes": {}
|
| 357 |
+
}
|
| 358 |
+
},
|
| 359 |
+
"total_flos": 8361496215552000.0,
|
| 360 |
+
"train_batch_size": 16,
|
| 361 |
+
"trial_name": null,
|
| 362 |
+
"trial_params": null
|
| 363 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e47fd9a569ce220c806b4a52069072c804efe830882485e57831413d7df87da6
|
| 3 |
+
size 5304
|
checkpoint-1000/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForQuestionAnswering"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"language": "english",
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"name": "Bert",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-1500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b12c9ba0433b464973b8793da675535099c25099968b1b8adc4e5cf39e06743f
|
| 3 |
+
size 435596088
|
checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:100cd0719d368c235075b060958929b9ff3bb258f59076131cddca626b155b9d
|
| 3 |
+
size 871311930
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba088525666769e1cf08d578b612b2c50f01fcf002579f70e2b63fb97fbe2652
|
| 3 |
+
size 14244
|
checkpoint-1500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ce1d988e39f6b2a2ca7c559657341b858aaf8b97ca115b786886db9286b7c34
|
| 3 |
+
size 988
|
checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c78ac9d3f464b1f309705681de3bb3d9ab7dd09b1112294e0b2f9bccb17181e8
|
| 3 |
+
size 1064
|
checkpoint-1500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-1500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,523 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1400,
|
| 3 |
+
"best_metric": 82.55212784144531,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.3973509933774834,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013245033112582781,
|
| 14 |
+
"grad_norm": 7.202547550201416,
|
| 15 |
+
"learning_rate": 9.998701986754968e-06,
|
| 16 |
+
"loss": 0.7094,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.026490066225165563,
|
| 21 |
+
"grad_norm": 9.662558555603027,
|
| 22 |
+
"learning_rate": 9.99737748344371e-06,
|
| 23 |
+
"loss": 0.68,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.026490066225165563,
|
| 28 |
+
"eval_HasAns_exact": 75.14792899408285,
|
| 29 |
+
"eval_HasAns_f1": 81.01957227969451,
|
| 30 |
+
"eval_HasAns_total": 15041,
|
| 31 |
+
"eval_best_exact": 75.14792899408285,
|
| 32 |
+
"eval_best_exact_thresh": 0.0,
|
| 33 |
+
"eval_best_f1": 81.01957227969451,
|
| 34 |
+
"eval_best_f1_thresh": 0.0,
|
| 35 |
+
"eval_exact": 75.14792899408285,
|
| 36 |
+
"eval_f1": 81.01957227969451,
|
| 37 |
+
"eval_loss": 0.5824803113937378,
|
| 38 |
+
"eval_runtime": 120.1578,
|
| 39 |
+
"eval_samples_per_second": 125.177,
|
| 40 |
+
"eval_steps_per_second": 7.831,
|
| 41 |
+
"eval_total": 15041,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.039735099337748346,
|
| 46 |
+
"grad_norm": 7.527211666107178,
|
| 47 |
+
"learning_rate": 9.996052980132451e-06,
|
| 48 |
+
"loss": 0.586,
|
| 49 |
+
"step": 150
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.052980132450331126,
|
| 53 |
+
"grad_norm": 5.419323921203613,
|
| 54 |
+
"learning_rate": 9.994728476821193e-06,
|
| 55 |
+
"loss": 0.6035,
|
| 56 |
+
"step": 200
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.052980132450331126,
|
| 60 |
+
"eval_HasAns_exact": 75.56013562927997,
|
| 61 |
+
"eval_HasAns_f1": 81.46153699536843,
|
| 62 |
+
"eval_HasAns_total": 15041,
|
| 63 |
+
"eval_best_exact": 75.56013562927997,
|
| 64 |
+
"eval_best_exact_thresh": 0.0,
|
| 65 |
+
"eval_best_f1": 81.46153699536843,
|
| 66 |
+
"eval_best_f1_thresh": 0.0,
|
| 67 |
+
"eval_exact": 75.56013562927997,
|
| 68 |
+
"eval_f1": 81.46153699536843,
|
| 69 |
+
"eval_loss": 0.5453085899353027,
|
| 70 |
+
"eval_runtime": 118.3719,
|
| 71 |
+
"eval_samples_per_second": 127.066,
|
| 72 |
+
"eval_steps_per_second": 7.95,
|
| 73 |
+
"eval_total": 15041,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.06622516556291391,
|
| 78 |
+
"grad_norm": 9.9099760055542,
|
| 79 |
+
"learning_rate": 9.993403973509935e-06,
|
| 80 |
+
"loss": 0.5616,
|
| 81 |
+
"step": 250
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.07947019867549669,
|
| 85 |
+
"grad_norm": 3.5184719562530518,
|
| 86 |
+
"learning_rate": 9.992079470198677e-06,
|
| 87 |
+
"loss": 0.5286,
|
| 88 |
+
"step": 300
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.07947019867549669,
|
| 92 |
+
"eval_HasAns_exact": 75.8526693703876,
|
| 93 |
+
"eval_HasAns_f1": 81.45745729278944,
|
| 94 |
+
"eval_HasAns_total": 15041,
|
| 95 |
+
"eval_best_exact": 75.8526693703876,
|
| 96 |
+
"eval_best_exact_thresh": 0.0,
|
| 97 |
+
"eval_best_f1": 81.45745729278944,
|
| 98 |
+
"eval_best_f1_thresh": 0.0,
|
| 99 |
+
"eval_exact": 75.8526693703876,
|
| 100 |
+
"eval_f1": 81.45745729278944,
|
| 101 |
+
"eval_loss": 0.5200360417366028,
|
| 102 |
+
"eval_runtime": 118.7441,
|
| 103 |
+
"eval_samples_per_second": 126.667,
|
| 104 |
+
"eval_steps_per_second": 7.925,
|
| 105 |
+
"eval_total": 15041,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.09271523178807947,
|
| 110 |
+
"grad_norm": 4.889588832855225,
|
| 111 |
+
"learning_rate": 9.990754966887418e-06,
|
| 112 |
+
"loss": 0.4943,
|
| 113 |
+
"step": 350
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.10596026490066225,
|
| 117 |
+
"grad_norm": 6.106881141662598,
|
| 118 |
+
"learning_rate": 9.98943046357616e-06,
|
| 119 |
+
"loss": 0.5957,
|
| 120 |
+
"step": 400
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.10596026490066225,
|
| 124 |
+
"eval_HasAns_exact": 76.39784588790639,
|
| 125 |
+
"eval_HasAns_f1": 81.91199737367272,
|
| 126 |
+
"eval_HasAns_total": 15041,
|
| 127 |
+
"eval_best_exact": 76.39784588790639,
|
| 128 |
+
"eval_best_exact_thresh": 0.0,
|
| 129 |
+
"eval_best_f1": 81.91199737367272,
|
| 130 |
+
"eval_best_f1_thresh": 0.0,
|
| 131 |
+
"eval_exact": 76.39784588790639,
|
| 132 |
+
"eval_f1": 81.91199737367272,
|
| 133 |
+
"eval_loss": 0.49782633781433105,
|
| 134 |
+
"eval_runtime": 118.2655,
|
| 135 |
+
"eval_samples_per_second": 127.18,
|
| 136 |
+
"eval_steps_per_second": 7.957,
|
| 137 |
+
"eval_total": 15041,
|
| 138 |
+
"step": 400
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.11920529801324503,
|
| 142 |
+
"grad_norm": 6.963775634765625,
|
| 143 |
+
"learning_rate": 9.988105960264902e-06,
|
| 144 |
+
"loss": 0.5675,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.13245033112582782,
|
| 149 |
+
"grad_norm": 7.559121608734131,
|
| 150 |
+
"learning_rate": 9.986781456953643e-06,
|
| 151 |
+
"loss": 0.5173,
|
| 152 |
+
"step": 500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.13245033112582782,
|
| 156 |
+
"eval_HasAns_exact": 75.94574828801277,
|
| 157 |
+
"eval_HasAns_f1": 81.43615773338746,
|
| 158 |
+
"eval_HasAns_total": 15041,
|
| 159 |
+
"eval_best_exact": 75.94574828801277,
|
| 160 |
+
"eval_best_exact_thresh": 0.0,
|
| 161 |
+
"eval_best_f1": 81.43615773338746,
|
| 162 |
+
"eval_best_f1_thresh": 0.0,
|
| 163 |
+
"eval_exact": 75.94574828801277,
|
| 164 |
+
"eval_f1": 81.43615773338746,
|
| 165 |
+
"eval_loss": 0.4918788969516754,
|
| 166 |
+
"eval_runtime": 118.4997,
|
| 167 |
+
"eval_samples_per_second": 126.929,
|
| 168 |
+
"eval_steps_per_second": 7.941,
|
| 169 |
+
"eval_total": 15041,
|
| 170 |
+
"step": 500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1456953642384106,
|
| 174 |
+
"grad_norm": 5.2273268699646,
|
| 175 |
+
"learning_rate": 9.985456953642385e-06,
|
| 176 |
+
"loss": 0.5059,
|
| 177 |
+
"step": 550
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.15894039735099338,
|
| 181 |
+
"grad_norm": 13.281010627746582,
|
| 182 |
+
"learning_rate": 9.984132450331127e-06,
|
| 183 |
+
"loss": 0.4847,
|
| 184 |
+
"step": 600
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.15894039735099338,
|
| 188 |
+
"eval_HasAns_exact": 76.42443986437073,
|
| 189 |
+
"eval_HasAns_f1": 81.98556495448281,
|
| 190 |
+
"eval_HasAns_total": 15041,
|
| 191 |
+
"eval_best_exact": 76.42443986437073,
|
| 192 |
+
"eval_best_exact_thresh": 0.0,
|
| 193 |
+
"eval_best_f1": 81.98556495448281,
|
| 194 |
+
"eval_best_f1_thresh": 0.0,
|
| 195 |
+
"eval_exact": 76.42443986437073,
|
| 196 |
+
"eval_f1": 81.98556495448281,
|
| 197 |
+
"eval_loss": 0.48446840047836304,
|
| 198 |
+
"eval_runtime": 118.3928,
|
| 199 |
+
"eval_samples_per_second": 127.043,
|
| 200 |
+
"eval_steps_per_second": 7.948,
|
| 201 |
+
"eval_total": 15041,
|
| 202 |
+
"step": 600
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.17218543046357615,
|
| 206 |
+
"grad_norm": 18.346189498901367,
|
| 207 |
+
"learning_rate": 9.982807947019869e-06,
|
| 208 |
+
"loss": 0.4857,
|
| 209 |
+
"step": 650
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.18543046357615894,
|
| 213 |
+
"grad_norm": 11.825461387634277,
|
| 214 |
+
"learning_rate": 9.98148344370861e-06,
|
| 215 |
+
"loss": 0.4764,
|
| 216 |
+
"step": 700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.18543046357615894,
|
| 220 |
+
"eval_HasAns_exact": 76.54411275846022,
|
| 221 |
+
"eval_HasAns_f1": 82.14175205695277,
|
| 222 |
+
"eval_HasAns_total": 15041,
|
| 223 |
+
"eval_best_exact": 76.54411275846022,
|
| 224 |
+
"eval_best_exact_thresh": 0.0,
|
| 225 |
+
"eval_best_f1": 82.14175205695277,
|
| 226 |
+
"eval_best_f1_thresh": 0.0,
|
| 227 |
+
"eval_exact": 76.54411275846022,
|
| 228 |
+
"eval_f1": 82.14175205695277,
|
| 229 |
+
"eval_loss": 0.4717336595058441,
|
| 230 |
+
"eval_runtime": 118.5022,
|
| 231 |
+
"eval_samples_per_second": 126.926,
|
| 232 |
+
"eval_steps_per_second": 7.941,
|
| 233 |
+
"eval_total": 15041,
|
| 234 |
+
"step": 700
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1986754966887417,
|
| 238 |
+
"grad_norm": 7.6593475341796875,
|
| 239 |
+
"learning_rate": 9.980158940397352e-06,
|
| 240 |
+
"loss": 0.486,
|
| 241 |
+
"step": 750
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2119205298013245,
|
| 245 |
+
"grad_norm": 6.091643333435059,
|
| 246 |
+
"learning_rate": 9.978834437086094e-06,
|
| 247 |
+
"loss": 0.4707,
|
| 248 |
+
"step": 800
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2119205298013245,
|
| 252 |
+
"eval_HasAns_exact": 76.86324047603217,
|
| 253 |
+
"eval_HasAns_f1": 82.2239211673303,
|
| 254 |
+
"eval_HasAns_total": 15041,
|
| 255 |
+
"eval_best_exact": 76.86324047603217,
|
| 256 |
+
"eval_best_exact_thresh": 0.0,
|
| 257 |
+
"eval_best_f1": 82.2239211673303,
|
| 258 |
+
"eval_best_f1_thresh": 0.0,
|
| 259 |
+
"eval_exact": 76.86324047603217,
|
| 260 |
+
"eval_f1": 82.2239211673303,
|
| 261 |
+
"eval_loss": 0.4601329565048218,
|
| 262 |
+
"eval_runtime": 119.1574,
|
| 263 |
+
"eval_samples_per_second": 126.228,
|
| 264 |
+
"eval_steps_per_second": 7.897,
|
| 265 |
+
"eval_total": 15041,
|
| 266 |
+
"step": 800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2251655629139073,
|
| 270 |
+
"grad_norm": 4.863158702850342,
|
| 271 |
+
"learning_rate": 9.977509933774835e-06,
|
| 272 |
+
"loss": 0.5042,
|
| 273 |
+
"step": 850
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.23841059602649006,
|
| 277 |
+
"grad_norm": 5.747822284698486,
|
| 278 |
+
"learning_rate": 9.976185430463577e-06,
|
| 279 |
+
"loss": 0.4725,
|
| 280 |
+
"step": 900
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.23841059602649006,
|
| 284 |
+
"eval_HasAns_exact": 76.98291337012166,
|
| 285 |
+
"eval_HasAns_f1": 82.39127215176644,
|
| 286 |
+
"eval_HasAns_total": 15041,
|
| 287 |
+
"eval_best_exact": 76.98291337012166,
|
| 288 |
+
"eval_best_exact_thresh": 0.0,
|
| 289 |
+
"eval_best_f1": 82.39127215176644,
|
| 290 |
+
"eval_best_f1_thresh": 0.0,
|
| 291 |
+
"eval_exact": 76.98291337012166,
|
| 292 |
+
"eval_f1": 82.39127215176644,
|
| 293 |
+
"eval_loss": 0.45768168568611145,
|
| 294 |
+
"eval_runtime": 118.7264,
|
| 295 |
+
"eval_samples_per_second": 126.686,
|
| 296 |
+
"eval_steps_per_second": 7.926,
|
| 297 |
+
"eval_total": 15041,
|
| 298 |
+
"step": 900
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.25165562913907286,
|
| 302 |
+
"grad_norm": 14.890397071838379,
|
| 303 |
+
"learning_rate": 9.974860927152319e-06,
|
| 304 |
+
"loss": 0.4476,
|
| 305 |
+
"step": 950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.26490066225165565,
|
| 309 |
+
"grad_norm": 8.837732315063477,
|
| 310 |
+
"learning_rate": 9.97353642384106e-06,
|
| 311 |
+
"loss": 0.4905,
|
| 312 |
+
"step": 1000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.26490066225165565,
|
| 316 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 317 |
+
"eval_HasAns_f1": 82.37799175072134,
|
| 318 |
+
"eval_HasAns_total": 15041,
|
| 319 |
+
"eval_best_exact": 77.10923475832723,
|
| 320 |
+
"eval_best_exact_thresh": 0.0,
|
| 321 |
+
"eval_best_f1": 82.37799175072134,
|
| 322 |
+
"eval_best_f1_thresh": 0.0,
|
| 323 |
+
"eval_exact": 77.10923475832723,
|
| 324 |
+
"eval_f1": 82.37799175072134,
|
| 325 |
+
"eval_loss": 0.44477367401123047,
|
| 326 |
+
"eval_runtime": 118.302,
|
| 327 |
+
"eval_samples_per_second": 127.141,
|
| 328 |
+
"eval_steps_per_second": 7.954,
|
| 329 |
+
"eval_total": 15041,
|
| 330 |
+
"step": 1000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.2781456953642384,
|
| 334 |
+
"grad_norm": 5.750672817230225,
|
| 335 |
+
"learning_rate": 9.972211920529802e-06,
|
| 336 |
+
"loss": 0.5041,
|
| 337 |
+
"step": 1050
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.2913907284768212,
|
| 341 |
+
"grad_norm": 5.720890998840332,
|
| 342 |
+
"learning_rate": 9.970887417218544e-06,
|
| 343 |
+
"loss": 0.5156,
|
| 344 |
+
"step": 1100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.2913907284768212,
|
| 348 |
+
"eval_HasAns_exact": 77.0826407818629,
|
| 349 |
+
"eval_HasAns_f1": 82.36916366838726,
|
| 350 |
+
"eval_HasAns_total": 15041,
|
| 351 |
+
"eval_best_exact": 77.0826407818629,
|
| 352 |
+
"eval_best_exact_thresh": 0.0,
|
| 353 |
+
"eval_best_f1": 82.36916366838726,
|
| 354 |
+
"eval_best_f1_thresh": 0.0,
|
| 355 |
+
"eval_exact": 77.0826407818629,
|
| 356 |
+
"eval_f1": 82.36916366838726,
|
| 357 |
+
"eval_loss": 0.43873801827430725,
|
| 358 |
+
"eval_runtime": 118.2856,
|
| 359 |
+
"eval_samples_per_second": 127.158,
|
| 360 |
+
"eval_steps_per_second": 7.955,
|
| 361 |
+
"eval_total": 15041,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.304635761589404,
|
| 366 |
+
"grad_norm": 6.990389823913574,
|
| 367 |
+
"learning_rate": 9.969562913907286e-06,
|
| 368 |
+
"loss": 0.4507,
|
| 369 |
+
"step": 1150
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.31788079470198677,
|
| 373 |
+
"grad_norm": 11.490405082702637,
|
| 374 |
+
"learning_rate": 9.968238410596027e-06,
|
| 375 |
+
"loss": 0.4482,
|
| 376 |
+
"step": 1200
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.31788079470198677,
|
| 380 |
+
"eval_HasAns_exact": 77.01615584070208,
|
| 381 |
+
"eval_HasAns_f1": 82.29906445927755,
|
| 382 |
+
"eval_HasAns_total": 15041,
|
| 383 |
+
"eval_best_exact": 77.01615584070208,
|
| 384 |
+
"eval_best_exact_thresh": 0.0,
|
| 385 |
+
"eval_best_f1": 82.29906445927755,
|
| 386 |
+
"eval_best_f1_thresh": 0.0,
|
| 387 |
+
"eval_exact": 77.01615584070208,
|
| 388 |
+
"eval_f1": 82.29906445927755,
|
| 389 |
+
"eval_loss": 0.42996492981910706,
|
| 390 |
+
"eval_runtime": 118.4966,
|
| 391 |
+
"eval_samples_per_second": 126.932,
|
| 392 |
+
"eval_steps_per_second": 7.941,
|
| 393 |
+
"eval_total": 15041,
|
| 394 |
+
"step": 1200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.33112582781456956,
|
| 398 |
+
"grad_norm": 2.8688676357269287,
|
| 399 |
+
"learning_rate": 9.966913907284769e-06,
|
| 400 |
+
"loss": 0.4593,
|
| 401 |
+
"step": 1250
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3443708609271523,
|
| 405 |
+
"grad_norm": 7.523802280426025,
|
| 406 |
+
"learning_rate": 9.96558940397351e-06,
|
| 407 |
+
"loss": 0.4548,
|
| 408 |
+
"step": 1300
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3443708609271523,
|
| 412 |
+
"eval_HasAns_exact": 76.94967089954126,
|
| 413 |
+
"eval_HasAns_f1": 82.17470826997452,
|
| 414 |
+
"eval_HasAns_total": 15041,
|
| 415 |
+
"eval_best_exact": 76.94967089954126,
|
| 416 |
+
"eval_best_exact_thresh": 0.0,
|
| 417 |
+
"eval_best_f1": 82.17470826997452,
|
| 418 |
+
"eval_best_f1_thresh": 0.0,
|
| 419 |
+
"eval_exact": 76.94967089954126,
|
| 420 |
+
"eval_f1": 82.17470826997452,
|
| 421 |
+
"eval_loss": 0.4349832832813263,
|
| 422 |
+
"eval_runtime": 118.6592,
|
| 423 |
+
"eval_samples_per_second": 126.758,
|
| 424 |
+
"eval_steps_per_second": 7.93,
|
| 425 |
+
"eval_total": 15041,
|
| 426 |
+
"step": 1300
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.3576158940397351,
|
| 430 |
+
"grad_norm": 10.102520942687988,
|
| 431 |
+
"learning_rate": 9.964264900662252e-06,
|
| 432 |
+
"loss": 0.4935,
|
| 433 |
+
"step": 1350
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.3708609271523179,
|
| 437 |
+
"grad_norm": 8.859588623046875,
|
| 438 |
+
"learning_rate": 9.962940397350994e-06,
|
| 439 |
+
"loss": 0.4344,
|
| 440 |
+
"step": 1400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.3708609271523179,
|
| 444 |
+
"eval_HasAns_exact": 77.19566518183632,
|
| 445 |
+
"eval_HasAns_f1": 82.55212784144531,
|
| 446 |
+
"eval_HasAns_total": 15041,
|
| 447 |
+
"eval_best_exact": 77.19566518183632,
|
| 448 |
+
"eval_best_exact_thresh": 0.0,
|
| 449 |
+
"eval_best_f1": 82.55212784144531,
|
| 450 |
+
"eval_best_f1_thresh": 0.0,
|
| 451 |
+
"eval_exact": 77.19566518183632,
|
| 452 |
+
"eval_f1": 82.55212784144531,
|
| 453 |
+
"eval_loss": 0.42366155982017517,
|
| 454 |
+
"eval_runtime": 119.1071,
|
| 455 |
+
"eval_samples_per_second": 126.281,
|
| 456 |
+
"eval_steps_per_second": 7.9,
|
| 457 |
+
"eval_total": 15041,
|
| 458 |
+
"step": 1400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3841059602649007,
|
| 462 |
+
"grad_norm": 5.980023384094238,
|
| 463 |
+
"learning_rate": 9.961615894039736e-06,
|
| 464 |
+
"loss": 0.4936,
|
| 465 |
+
"step": 1450
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3973509933774834,
|
| 469 |
+
"grad_norm": 5.025954246520996,
|
| 470 |
+
"learning_rate": 9.960291390728478e-06,
|
| 471 |
+
"loss": 0.4431,
|
| 472 |
+
"step": 1500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.3973509933774834,
|
| 476 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 477 |
+
"eval_HasAns_f1": 82.41666220398032,
|
| 478 |
+
"eval_HasAns_total": 15041,
|
| 479 |
+
"eval_best_exact": 77.10923475832723,
|
| 480 |
+
"eval_best_exact_thresh": 0.0,
|
| 481 |
+
"eval_best_f1": 82.41666220398032,
|
| 482 |
+
"eval_best_f1_thresh": 0.0,
|
| 483 |
+
"eval_exact": 77.10923475832723,
|
| 484 |
+
"eval_f1": 82.41666220398032,
|
| 485 |
+
"eval_loss": 0.4237441122531891,
|
| 486 |
+
"eval_runtime": 118.7094,
|
| 487 |
+
"eval_samples_per_second": 126.704,
|
| 488 |
+
"eval_steps_per_second": 7.927,
|
| 489 |
+
"eval_total": 15041,
|
| 490 |
+
"step": 1500
|
| 491 |
+
}
|
| 492 |
+
],
|
| 493 |
+
"logging_steps": 50,
|
| 494 |
+
"max_steps": 377500,
|
| 495 |
+
"num_input_tokens_seen": 0,
|
| 496 |
+
"num_train_epochs": 100,
|
| 497 |
+
"save_steps": 500,
|
| 498 |
+
"stateful_callbacks": {
|
| 499 |
+
"EarlyStoppingCallback": {
|
| 500 |
+
"args": {
|
| 501 |
+
"early_stopping_patience": 10,
|
| 502 |
+
"early_stopping_threshold": 0.0
|
| 503 |
+
},
|
| 504 |
+
"attributes": {
|
| 505 |
+
"early_stopping_patience_counter": 1
|
| 506 |
+
}
|
| 507 |
+
},
|
| 508 |
+
"TrainerControl": {
|
| 509 |
+
"args": {
|
| 510 |
+
"should_epoch_stop": false,
|
| 511 |
+
"should_evaluate": false,
|
| 512 |
+
"should_log": false,
|
| 513 |
+
"should_save": true,
|
| 514 |
+
"should_training_stop": false
|
| 515 |
+
},
|
| 516 |
+
"attributes": {}
|
| 517 |
+
}
|
| 518 |
+
},
|
| 519 |
+
"total_flos": 1.2542244323328e+16,
|
| 520 |
+
"train_batch_size": 16,
|
| 521 |
+
"trial_name": null,
|
| 522 |
+
"trial_params": null
|
| 523 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e47fd9a569ce220c806b4a52069072c804efe830882485e57831413d7df87da6
|
| 3 |
+
size 5304
|
checkpoint-1500/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForQuestionAnswering"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"language": "english",
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"name": "Bert",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-2000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77c818f37165024ec341b1d3e78509efb565906b124ec375f9b410be1750e2b0
|
| 3 |
+
size 435596088
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc06d113d216e4a8bd338c02165a4cc2155be3914ce4b86eade5497a73aa7d1e
|
| 3 |
+
size 871311930
|
checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c739be83fa1f3a50eced1baaf28a6293b7420251b4834aa76b330fef386f8a43
|
| 3 |
+
size 14244
|
checkpoint-2000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:171dd51250bdb9db45bc30d0105e09652815ee6f81e02e37202c3cda8e3aa351
|
| 3 |
+
size 988
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ed85419d815bc7f6be9bb18f1acdbbf742ddefb148e62ed1628c911159001c4
|
| 3 |
+
size 1064
|
checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,683 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1900,
|
| 3 |
+
"best_metric": 82.9212903481507,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.5298013245033113,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013245033112582781,
|
| 14 |
+
"grad_norm": 7.202547550201416,
|
| 15 |
+
"learning_rate": 9.998701986754968e-06,
|
| 16 |
+
"loss": 0.7094,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.026490066225165563,
|
| 21 |
+
"grad_norm": 9.662558555603027,
|
| 22 |
+
"learning_rate": 9.99737748344371e-06,
|
| 23 |
+
"loss": 0.68,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.026490066225165563,
|
| 28 |
+
"eval_HasAns_exact": 75.14792899408285,
|
| 29 |
+
"eval_HasAns_f1": 81.01957227969451,
|
| 30 |
+
"eval_HasAns_total": 15041,
|
| 31 |
+
"eval_best_exact": 75.14792899408285,
|
| 32 |
+
"eval_best_exact_thresh": 0.0,
|
| 33 |
+
"eval_best_f1": 81.01957227969451,
|
| 34 |
+
"eval_best_f1_thresh": 0.0,
|
| 35 |
+
"eval_exact": 75.14792899408285,
|
| 36 |
+
"eval_f1": 81.01957227969451,
|
| 37 |
+
"eval_loss": 0.5824803113937378,
|
| 38 |
+
"eval_runtime": 120.1578,
|
| 39 |
+
"eval_samples_per_second": 125.177,
|
| 40 |
+
"eval_steps_per_second": 7.831,
|
| 41 |
+
"eval_total": 15041,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.039735099337748346,
|
| 46 |
+
"grad_norm": 7.527211666107178,
|
| 47 |
+
"learning_rate": 9.996052980132451e-06,
|
| 48 |
+
"loss": 0.586,
|
| 49 |
+
"step": 150
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.052980132450331126,
|
| 53 |
+
"grad_norm": 5.419323921203613,
|
| 54 |
+
"learning_rate": 9.994728476821193e-06,
|
| 55 |
+
"loss": 0.6035,
|
| 56 |
+
"step": 200
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.052980132450331126,
|
| 60 |
+
"eval_HasAns_exact": 75.56013562927997,
|
| 61 |
+
"eval_HasAns_f1": 81.46153699536843,
|
| 62 |
+
"eval_HasAns_total": 15041,
|
| 63 |
+
"eval_best_exact": 75.56013562927997,
|
| 64 |
+
"eval_best_exact_thresh": 0.0,
|
| 65 |
+
"eval_best_f1": 81.46153699536843,
|
| 66 |
+
"eval_best_f1_thresh": 0.0,
|
| 67 |
+
"eval_exact": 75.56013562927997,
|
| 68 |
+
"eval_f1": 81.46153699536843,
|
| 69 |
+
"eval_loss": 0.5453085899353027,
|
| 70 |
+
"eval_runtime": 118.3719,
|
| 71 |
+
"eval_samples_per_second": 127.066,
|
| 72 |
+
"eval_steps_per_second": 7.95,
|
| 73 |
+
"eval_total": 15041,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.06622516556291391,
|
| 78 |
+
"grad_norm": 9.9099760055542,
|
| 79 |
+
"learning_rate": 9.993403973509935e-06,
|
| 80 |
+
"loss": 0.5616,
|
| 81 |
+
"step": 250
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.07947019867549669,
|
| 85 |
+
"grad_norm": 3.5184719562530518,
|
| 86 |
+
"learning_rate": 9.992079470198677e-06,
|
| 87 |
+
"loss": 0.5286,
|
| 88 |
+
"step": 300
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.07947019867549669,
|
| 92 |
+
"eval_HasAns_exact": 75.8526693703876,
|
| 93 |
+
"eval_HasAns_f1": 81.45745729278944,
|
| 94 |
+
"eval_HasAns_total": 15041,
|
| 95 |
+
"eval_best_exact": 75.8526693703876,
|
| 96 |
+
"eval_best_exact_thresh": 0.0,
|
| 97 |
+
"eval_best_f1": 81.45745729278944,
|
| 98 |
+
"eval_best_f1_thresh": 0.0,
|
| 99 |
+
"eval_exact": 75.8526693703876,
|
| 100 |
+
"eval_f1": 81.45745729278944,
|
| 101 |
+
"eval_loss": 0.5200360417366028,
|
| 102 |
+
"eval_runtime": 118.7441,
|
| 103 |
+
"eval_samples_per_second": 126.667,
|
| 104 |
+
"eval_steps_per_second": 7.925,
|
| 105 |
+
"eval_total": 15041,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.09271523178807947,
|
| 110 |
+
"grad_norm": 4.889588832855225,
|
| 111 |
+
"learning_rate": 9.990754966887418e-06,
|
| 112 |
+
"loss": 0.4943,
|
| 113 |
+
"step": 350
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.10596026490066225,
|
| 117 |
+
"grad_norm": 6.106881141662598,
|
| 118 |
+
"learning_rate": 9.98943046357616e-06,
|
| 119 |
+
"loss": 0.5957,
|
| 120 |
+
"step": 400
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.10596026490066225,
|
| 124 |
+
"eval_HasAns_exact": 76.39784588790639,
|
| 125 |
+
"eval_HasAns_f1": 81.91199737367272,
|
| 126 |
+
"eval_HasAns_total": 15041,
|
| 127 |
+
"eval_best_exact": 76.39784588790639,
|
| 128 |
+
"eval_best_exact_thresh": 0.0,
|
| 129 |
+
"eval_best_f1": 81.91199737367272,
|
| 130 |
+
"eval_best_f1_thresh": 0.0,
|
| 131 |
+
"eval_exact": 76.39784588790639,
|
| 132 |
+
"eval_f1": 81.91199737367272,
|
| 133 |
+
"eval_loss": 0.49782633781433105,
|
| 134 |
+
"eval_runtime": 118.2655,
|
| 135 |
+
"eval_samples_per_second": 127.18,
|
| 136 |
+
"eval_steps_per_second": 7.957,
|
| 137 |
+
"eval_total": 15041,
|
| 138 |
+
"step": 400
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.11920529801324503,
|
| 142 |
+
"grad_norm": 6.963775634765625,
|
| 143 |
+
"learning_rate": 9.988105960264902e-06,
|
| 144 |
+
"loss": 0.5675,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.13245033112582782,
|
| 149 |
+
"grad_norm": 7.559121608734131,
|
| 150 |
+
"learning_rate": 9.986781456953643e-06,
|
| 151 |
+
"loss": 0.5173,
|
| 152 |
+
"step": 500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.13245033112582782,
|
| 156 |
+
"eval_HasAns_exact": 75.94574828801277,
|
| 157 |
+
"eval_HasAns_f1": 81.43615773338746,
|
| 158 |
+
"eval_HasAns_total": 15041,
|
| 159 |
+
"eval_best_exact": 75.94574828801277,
|
| 160 |
+
"eval_best_exact_thresh": 0.0,
|
| 161 |
+
"eval_best_f1": 81.43615773338746,
|
| 162 |
+
"eval_best_f1_thresh": 0.0,
|
| 163 |
+
"eval_exact": 75.94574828801277,
|
| 164 |
+
"eval_f1": 81.43615773338746,
|
| 165 |
+
"eval_loss": 0.4918788969516754,
|
| 166 |
+
"eval_runtime": 118.4997,
|
| 167 |
+
"eval_samples_per_second": 126.929,
|
| 168 |
+
"eval_steps_per_second": 7.941,
|
| 169 |
+
"eval_total": 15041,
|
| 170 |
+
"step": 500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1456953642384106,
|
| 174 |
+
"grad_norm": 5.2273268699646,
|
| 175 |
+
"learning_rate": 9.985456953642385e-06,
|
| 176 |
+
"loss": 0.5059,
|
| 177 |
+
"step": 550
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.15894039735099338,
|
| 181 |
+
"grad_norm": 13.281010627746582,
|
| 182 |
+
"learning_rate": 9.984132450331127e-06,
|
| 183 |
+
"loss": 0.4847,
|
| 184 |
+
"step": 600
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.15894039735099338,
|
| 188 |
+
"eval_HasAns_exact": 76.42443986437073,
|
| 189 |
+
"eval_HasAns_f1": 81.98556495448281,
|
| 190 |
+
"eval_HasAns_total": 15041,
|
| 191 |
+
"eval_best_exact": 76.42443986437073,
|
| 192 |
+
"eval_best_exact_thresh": 0.0,
|
| 193 |
+
"eval_best_f1": 81.98556495448281,
|
| 194 |
+
"eval_best_f1_thresh": 0.0,
|
| 195 |
+
"eval_exact": 76.42443986437073,
|
| 196 |
+
"eval_f1": 81.98556495448281,
|
| 197 |
+
"eval_loss": 0.48446840047836304,
|
| 198 |
+
"eval_runtime": 118.3928,
|
| 199 |
+
"eval_samples_per_second": 127.043,
|
| 200 |
+
"eval_steps_per_second": 7.948,
|
| 201 |
+
"eval_total": 15041,
|
| 202 |
+
"step": 600
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.17218543046357615,
|
| 206 |
+
"grad_norm": 18.346189498901367,
|
| 207 |
+
"learning_rate": 9.982807947019869e-06,
|
| 208 |
+
"loss": 0.4857,
|
| 209 |
+
"step": 650
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.18543046357615894,
|
| 213 |
+
"grad_norm": 11.825461387634277,
|
| 214 |
+
"learning_rate": 9.98148344370861e-06,
|
| 215 |
+
"loss": 0.4764,
|
| 216 |
+
"step": 700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.18543046357615894,
|
| 220 |
+
"eval_HasAns_exact": 76.54411275846022,
|
| 221 |
+
"eval_HasAns_f1": 82.14175205695277,
|
| 222 |
+
"eval_HasAns_total": 15041,
|
| 223 |
+
"eval_best_exact": 76.54411275846022,
|
| 224 |
+
"eval_best_exact_thresh": 0.0,
|
| 225 |
+
"eval_best_f1": 82.14175205695277,
|
| 226 |
+
"eval_best_f1_thresh": 0.0,
|
| 227 |
+
"eval_exact": 76.54411275846022,
|
| 228 |
+
"eval_f1": 82.14175205695277,
|
| 229 |
+
"eval_loss": 0.4717336595058441,
|
| 230 |
+
"eval_runtime": 118.5022,
|
| 231 |
+
"eval_samples_per_second": 126.926,
|
| 232 |
+
"eval_steps_per_second": 7.941,
|
| 233 |
+
"eval_total": 15041,
|
| 234 |
+
"step": 700
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1986754966887417,
|
| 238 |
+
"grad_norm": 7.6593475341796875,
|
| 239 |
+
"learning_rate": 9.980158940397352e-06,
|
| 240 |
+
"loss": 0.486,
|
| 241 |
+
"step": 750
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2119205298013245,
|
| 245 |
+
"grad_norm": 6.091643333435059,
|
| 246 |
+
"learning_rate": 9.978834437086094e-06,
|
| 247 |
+
"loss": 0.4707,
|
| 248 |
+
"step": 800
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2119205298013245,
|
| 252 |
+
"eval_HasAns_exact": 76.86324047603217,
|
| 253 |
+
"eval_HasAns_f1": 82.2239211673303,
|
| 254 |
+
"eval_HasAns_total": 15041,
|
| 255 |
+
"eval_best_exact": 76.86324047603217,
|
| 256 |
+
"eval_best_exact_thresh": 0.0,
|
| 257 |
+
"eval_best_f1": 82.2239211673303,
|
| 258 |
+
"eval_best_f1_thresh": 0.0,
|
| 259 |
+
"eval_exact": 76.86324047603217,
|
| 260 |
+
"eval_f1": 82.2239211673303,
|
| 261 |
+
"eval_loss": 0.4601329565048218,
|
| 262 |
+
"eval_runtime": 119.1574,
|
| 263 |
+
"eval_samples_per_second": 126.228,
|
| 264 |
+
"eval_steps_per_second": 7.897,
|
| 265 |
+
"eval_total": 15041,
|
| 266 |
+
"step": 800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2251655629139073,
|
| 270 |
+
"grad_norm": 4.863158702850342,
|
| 271 |
+
"learning_rate": 9.977509933774835e-06,
|
| 272 |
+
"loss": 0.5042,
|
| 273 |
+
"step": 850
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.23841059602649006,
|
| 277 |
+
"grad_norm": 5.747822284698486,
|
| 278 |
+
"learning_rate": 9.976185430463577e-06,
|
| 279 |
+
"loss": 0.4725,
|
| 280 |
+
"step": 900
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.23841059602649006,
|
| 284 |
+
"eval_HasAns_exact": 76.98291337012166,
|
| 285 |
+
"eval_HasAns_f1": 82.39127215176644,
|
| 286 |
+
"eval_HasAns_total": 15041,
|
| 287 |
+
"eval_best_exact": 76.98291337012166,
|
| 288 |
+
"eval_best_exact_thresh": 0.0,
|
| 289 |
+
"eval_best_f1": 82.39127215176644,
|
| 290 |
+
"eval_best_f1_thresh": 0.0,
|
| 291 |
+
"eval_exact": 76.98291337012166,
|
| 292 |
+
"eval_f1": 82.39127215176644,
|
| 293 |
+
"eval_loss": 0.45768168568611145,
|
| 294 |
+
"eval_runtime": 118.7264,
|
| 295 |
+
"eval_samples_per_second": 126.686,
|
| 296 |
+
"eval_steps_per_second": 7.926,
|
| 297 |
+
"eval_total": 15041,
|
| 298 |
+
"step": 900
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.25165562913907286,
|
| 302 |
+
"grad_norm": 14.890397071838379,
|
| 303 |
+
"learning_rate": 9.974860927152319e-06,
|
| 304 |
+
"loss": 0.4476,
|
| 305 |
+
"step": 950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.26490066225165565,
|
| 309 |
+
"grad_norm": 8.837732315063477,
|
| 310 |
+
"learning_rate": 9.97353642384106e-06,
|
| 311 |
+
"loss": 0.4905,
|
| 312 |
+
"step": 1000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.26490066225165565,
|
| 316 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 317 |
+
"eval_HasAns_f1": 82.37799175072134,
|
| 318 |
+
"eval_HasAns_total": 15041,
|
| 319 |
+
"eval_best_exact": 77.10923475832723,
|
| 320 |
+
"eval_best_exact_thresh": 0.0,
|
| 321 |
+
"eval_best_f1": 82.37799175072134,
|
| 322 |
+
"eval_best_f1_thresh": 0.0,
|
| 323 |
+
"eval_exact": 77.10923475832723,
|
| 324 |
+
"eval_f1": 82.37799175072134,
|
| 325 |
+
"eval_loss": 0.44477367401123047,
|
| 326 |
+
"eval_runtime": 118.302,
|
| 327 |
+
"eval_samples_per_second": 127.141,
|
| 328 |
+
"eval_steps_per_second": 7.954,
|
| 329 |
+
"eval_total": 15041,
|
| 330 |
+
"step": 1000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.2781456953642384,
|
| 334 |
+
"grad_norm": 5.750672817230225,
|
| 335 |
+
"learning_rate": 9.972211920529802e-06,
|
| 336 |
+
"loss": 0.5041,
|
| 337 |
+
"step": 1050
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.2913907284768212,
|
| 341 |
+
"grad_norm": 5.720890998840332,
|
| 342 |
+
"learning_rate": 9.970887417218544e-06,
|
| 343 |
+
"loss": 0.5156,
|
| 344 |
+
"step": 1100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.2913907284768212,
|
| 348 |
+
"eval_HasAns_exact": 77.0826407818629,
|
| 349 |
+
"eval_HasAns_f1": 82.36916366838726,
|
| 350 |
+
"eval_HasAns_total": 15041,
|
| 351 |
+
"eval_best_exact": 77.0826407818629,
|
| 352 |
+
"eval_best_exact_thresh": 0.0,
|
| 353 |
+
"eval_best_f1": 82.36916366838726,
|
| 354 |
+
"eval_best_f1_thresh": 0.0,
|
| 355 |
+
"eval_exact": 77.0826407818629,
|
| 356 |
+
"eval_f1": 82.36916366838726,
|
| 357 |
+
"eval_loss": 0.43873801827430725,
|
| 358 |
+
"eval_runtime": 118.2856,
|
| 359 |
+
"eval_samples_per_second": 127.158,
|
| 360 |
+
"eval_steps_per_second": 7.955,
|
| 361 |
+
"eval_total": 15041,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.304635761589404,
|
| 366 |
+
"grad_norm": 6.990389823913574,
|
| 367 |
+
"learning_rate": 9.969562913907286e-06,
|
| 368 |
+
"loss": 0.4507,
|
| 369 |
+
"step": 1150
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.31788079470198677,
|
| 373 |
+
"grad_norm": 11.490405082702637,
|
| 374 |
+
"learning_rate": 9.968238410596027e-06,
|
| 375 |
+
"loss": 0.4482,
|
| 376 |
+
"step": 1200
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.31788079470198677,
|
| 380 |
+
"eval_HasAns_exact": 77.01615584070208,
|
| 381 |
+
"eval_HasAns_f1": 82.29906445927755,
|
| 382 |
+
"eval_HasAns_total": 15041,
|
| 383 |
+
"eval_best_exact": 77.01615584070208,
|
| 384 |
+
"eval_best_exact_thresh": 0.0,
|
| 385 |
+
"eval_best_f1": 82.29906445927755,
|
| 386 |
+
"eval_best_f1_thresh": 0.0,
|
| 387 |
+
"eval_exact": 77.01615584070208,
|
| 388 |
+
"eval_f1": 82.29906445927755,
|
| 389 |
+
"eval_loss": 0.42996492981910706,
|
| 390 |
+
"eval_runtime": 118.4966,
|
| 391 |
+
"eval_samples_per_second": 126.932,
|
| 392 |
+
"eval_steps_per_second": 7.941,
|
| 393 |
+
"eval_total": 15041,
|
| 394 |
+
"step": 1200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.33112582781456956,
|
| 398 |
+
"grad_norm": 2.8688676357269287,
|
| 399 |
+
"learning_rate": 9.966913907284769e-06,
|
| 400 |
+
"loss": 0.4593,
|
| 401 |
+
"step": 1250
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3443708609271523,
|
| 405 |
+
"grad_norm": 7.523802280426025,
|
| 406 |
+
"learning_rate": 9.96558940397351e-06,
|
| 407 |
+
"loss": 0.4548,
|
| 408 |
+
"step": 1300
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3443708609271523,
|
| 412 |
+
"eval_HasAns_exact": 76.94967089954126,
|
| 413 |
+
"eval_HasAns_f1": 82.17470826997452,
|
| 414 |
+
"eval_HasAns_total": 15041,
|
| 415 |
+
"eval_best_exact": 76.94967089954126,
|
| 416 |
+
"eval_best_exact_thresh": 0.0,
|
| 417 |
+
"eval_best_f1": 82.17470826997452,
|
| 418 |
+
"eval_best_f1_thresh": 0.0,
|
| 419 |
+
"eval_exact": 76.94967089954126,
|
| 420 |
+
"eval_f1": 82.17470826997452,
|
| 421 |
+
"eval_loss": 0.4349832832813263,
|
| 422 |
+
"eval_runtime": 118.6592,
|
| 423 |
+
"eval_samples_per_second": 126.758,
|
| 424 |
+
"eval_steps_per_second": 7.93,
|
| 425 |
+
"eval_total": 15041,
|
| 426 |
+
"step": 1300
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.3576158940397351,
|
| 430 |
+
"grad_norm": 10.102520942687988,
|
| 431 |
+
"learning_rate": 9.964264900662252e-06,
|
| 432 |
+
"loss": 0.4935,
|
| 433 |
+
"step": 1350
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.3708609271523179,
|
| 437 |
+
"grad_norm": 8.859588623046875,
|
| 438 |
+
"learning_rate": 9.962940397350994e-06,
|
| 439 |
+
"loss": 0.4344,
|
| 440 |
+
"step": 1400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.3708609271523179,
|
| 444 |
+
"eval_HasAns_exact": 77.19566518183632,
|
| 445 |
+
"eval_HasAns_f1": 82.55212784144531,
|
| 446 |
+
"eval_HasAns_total": 15041,
|
| 447 |
+
"eval_best_exact": 77.19566518183632,
|
| 448 |
+
"eval_best_exact_thresh": 0.0,
|
| 449 |
+
"eval_best_f1": 82.55212784144531,
|
| 450 |
+
"eval_best_f1_thresh": 0.0,
|
| 451 |
+
"eval_exact": 77.19566518183632,
|
| 452 |
+
"eval_f1": 82.55212784144531,
|
| 453 |
+
"eval_loss": 0.42366155982017517,
|
| 454 |
+
"eval_runtime": 119.1071,
|
| 455 |
+
"eval_samples_per_second": 126.281,
|
| 456 |
+
"eval_steps_per_second": 7.9,
|
| 457 |
+
"eval_total": 15041,
|
| 458 |
+
"step": 1400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3841059602649007,
|
| 462 |
+
"grad_norm": 5.980023384094238,
|
| 463 |
+
"learning_rate": 9.961615894039736e-06,
|
| 464 |
+
"loss": 0.4936,
|
| 465 |
+
"step": 1450
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3973509933774834,
|
| 469 |
+
"grad_norm": 5.025954246520996,
|
| 470 |
+
"learning_rate": 9.960291390728478e-06,
|
| 471 |
+
"loss": 0.4431,
|
| 472 |
+
"step": 1500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.3973509933774834,
|
| 476 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 477 |
+
"eval_HasAns_f1": 82.41666220398032,
|
| 478 |
+
"eval_HasAns_total": 15041,
|
| 479 |
+
"eval_best_exact": 77.10923475832723,
|
| 480 |
+
"eval_best_exact_thresh": 0.0,
|
| 481 |
+
"eval_best_f1": 82.41666220398032,
|
| 482 |
+
"eval_best_f1_thresh": 0.0,
|
| 483 |
+
"eval_exact": 77.10923475832723,
|
| 484 |
+
"eval_f1": 82.41666220398032,
|
| 485 |
+
"eval_loss": 0.4237441122531891,
|
| 486 |
+
"eval_runtime": 118.7094,
|
| 487 |
+
"eval_samples_per_second": 126.704,
|
| 488 |
+
"eval_steps_per_second": 7.927,
|
| 489 |
+
"eval_total": 15041,
|
| 490 |
+
"step": 1500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.4105960264900662,
|
| 494 |
+
"grad_norm": 13.039314270019531,
|
| 495 |
+
"learning_rate": 9.95896688741722e-06,
|
| 496 |
+
"loss": 0.4907,
|
| 497 |
+
"step": 1550
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.423841059602649,
|
| 501 |
+
"grad_norm": 8.910651206970215,
|
| 502 |
+
"learning_rate": 9.957642384105961e-06,
|
| 503 |
+
"loss": 0.4462,
|
| 504 |
+
"step": 1600
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.423841059602649,
|
| 508 |
+
"eval_HasAns_exact": 77.27544711122931,
|
| 509 |
+
"eval_HasAns_f1": 82.45441363866084,
|
| 510 |
+
"eval_HasAns_total": 15041,
|
| 511 |
+
"eval_best_exact": 77.27544711122931,
|
| 512 |
+
"eval_best_exact_thresh": 0.0,
|
| 513 |
+
"eval_best_f1": 82.45441363866084,
|
| 514 |
+
"eval_best_f1_thresh": 0.0,
|
| 515 |
+
"eval_exact": 77.27544711122931,
|
| 516 |
+
"eval_f1": 82.45441363866084,
|
| 517 |
+
"eval_loss": 0.4120015501976013,
|
| 518 |
+
"eval_runtime": 118.2118,
|
| 519 |
+
"eval_samples_per_second": 127.238,
|
| 520 |
+
"eval_steps_per_second": 7.96,
|
| 521 |
+
"eval_total": 15041,
|
| 522 |
+
"step": 1600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.4370860927152318,
|
| 526 |
+
"grad_norm": 7.536431312561035,
|
| 527 |
+
"learning_rate": 9.956317880794703e-06,
|
| 528 |
+
"loss": 0.423,
|
| 529 |
+
"step": 1650
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.4503311258278146,
|
| 533 |
+
"grad_norm": 6.14304780960083,
|
| 534 |
+
"learning_rate": 9.954993377483444e-06,
|
| 535 |
+
"loss": 0.4353,
|
| 536 |
+
"step": 1700
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.4503311258278146,
|
| 540 |
+
"eval_HasAns_exact": 77.64111428761386,
|
| 541 |
+
"eval_HasAns_f1": 82.82953930481963,
|
| 542 |
+
"eval_HasAns_total": 15041,
|
| 543 |
+
"eval_best_exact": 77.64111428761386,
|
| 544 |
+
"eval_best_exact_thresh": 0.0,
|
| 545 |
+
"eval_best_f1": 82.82953930481963,
|
| 546 |
+
"eval_best_f1_thresh": 0.0,
|
| 547 |
+
"eval_exact": 77.64111428761386,
|
| 548 |
+
"eval_f1": 82.82953930481963,
|
| 549 |
+
"eval_loss": 0.4078502357006073,
|
| 550 |
+
"eval_runtime": 118.4213,
|
| 551 |
+
"eval_samples_per_second": 127.013,
|
| 552 |
+
"eval_steps_per_second": 7.946,
|
| 553 |
+
"eval_total": 15041,
|
| 554 |
+
"step": 1700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.46357615894039733,
|
| 558 |
+
"grad_norm": 7.738461017608643,
|
| 559 |
+
"learning_rate": 9.953668874172186e-06,
|
| 560 |
+
"loss": 0.4166,
|
| 561 |
+
"step": 1750
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.4768211920529801,
|
| 565 |
+
"grad_norm": 5.486303329467773,
|
| 566 |
+
"learning_rate": 9.952344370860928e-06,
|
| 567 |
+
"loss": 0.4344,
|
| 568 |
+
"step": 1800
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.4768211920529801,
|
| 572 |
+
"eval_HasAns_exact": 77.53473838175653,
|
| 573 |
+
"eval_HasAns_f1": 82.83341057132324,
|
| 574 |
+
"eval_HasAns_total": 15041,
|
| 575 |
+
"eval_best_exact": 77.53473838175653,
|
| 576 |
+
"eval_best_exact_thresh": 0.0,
|
| 577 |
+
"eval_best_f1": 82.83341057132324,
|
| 578 |
+
"eval_best_f1_thresh": 0.0,
|
| 579 |
+
"eval_exact": 77.53473838175653,
|
| 580 |
+
"eval_f1": 82.83341057132324,
|
| 581 |
+
"eval_loss": 0.4005696475505829,
|
| 582 |
+
"eval_runtime": 118.4606,
|
| 583 |
+
"eval_samples_per_second": 126.97,
|
| 584 |
+
"eval_steps_per_second": 7.944,
|
| 585 |
+
"eval_total": 15041,
|
| 586 |
+
"step": 1800
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4900662251655629,
|
| 590 |
+
"grad_norm": 11.653322219848633,
|
| 591 |
+
"learning_rate": 9.95101986754967e-06,
|
| 592 |
+
"loss": 0.444,
|
| 593 |
+
"step": 1850
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.5033112582781457,
|
| 597 |
+
"grad_norm": 9.203779220581055,
|
| 598 |
+
"learning_rate": 9.949695364238411e-06,
|
| 599 |
+
"loss": 0.3835,
|
| 600 |
+
"step": 1900
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.5033112582781457,
|
| 604 |
+
"eval_HasAns_exact": 77.813975134632,
|
| 605 |
+
"eval_HasAns_f1": 82.9212903481507,
|
| 606 |
+
"eval_HasAns_total": 15041,
|
| 607 |
+
"eval_best_exact": 77.813975134632,
|
| 608 |
+
"eval_best_exact_thresh": 0.0,
|
| 609 |
+
"eval_best_f1": 82.9212903481507,
|
| 610 |
+
"eval_best_f1_thresh": 0.0,
|
| 611 |
+
"eval_exact": 77.813975134632,
|
| 612 |
+
"eval_f1": 82.9212903481507,
|
| 613 |
+
"eval_loss": 0.40120014548301697,
|
| 614 |
+
"eval_runtime": 118.8575,
|
| 615 |
+
"eval_samples_per_second": 126.547,
|
| 616 |
+
"eval_steps_per_second": 7.917,
|
| 617 |
+
"eval_total": 15041,
|
| 618 |
+
"step": 1900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5165562913907285,
|
| 622 |
+
"grad_norm": 11.891190528869629,
|
| 623 |
+
"learning_rate": 9.948370860927153e-06,
|
| 624 |
+
"loss": 0.4682,
|
| 625 |
+
"step": 1950
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5298013245033113,
|
| 629 |
+
"grad_norm": 7.65544319152832,
|
| 630 |
+
"learning_rate": 9.947046357615895e-06,
|
| 631 |
+
"loss": 0.4618,
|
| 632 |
+
"step": 2000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5298013245033113,
|
| 636 |
+
"eval_HasAns_exact": 77.63446579349777,
|
| 637 |
+
"eval_HasAns_f1": 82.84673293378647,
|
| 638 |
+
"eval_HasAns_total": 15041,
|
| 639 |
+
"eval_best_exact": 77.63446579349777,
|
| 640 |
+
"eval_best_exact_thresh": 0.0,
|
| 641 |
+
"eval_best_f1": 82.84673293378647,
|
| 642 |
+
"eval_best_f1_thresh": 0.0,
|
| 643 |
+
"eval_exact": 77.63446579349777,
|
| 644 |
+
"eval_f1": 82.84673293378647,
|
| 645 |
+
"eval_loss": 0.38913631439208984,
|
| 646 |
+
"eval_runtime": 118.6931,
|
| 647 |
+
"eval_samples_per_second": 126.722,
|
| 648 |
+
"eval_steps_per_second": 7.928,
|
| 649 |
+
"eval_total": 15041,
|
| 650 |
+
"step": 2000
|
| 651 |
+
}
|
| 652 |
+
],
|
| 653 |
+
"logging_steps": 50,
|
| 654 |
+
"max_steps": 377500,
|
| 655 |
+
"num_input_tokens_seen": 0,
|
| 656 |
+
"num_train_epochs": 100,
|
| 657 |
+
"save_steps": 500,
|
| 658 |
+
"stateful_callbacks": {
|
| 659 |
+
"EarlyStoppingCallback": {
|
| 660 |
+
"args": {
|
| 661 |
+
"early_stopping_patience": 10,
|
| 662 |
+
"early_stopping_threshold": 0.0
|
| 663 |
+
},
|
| 664 |
+
"attributes": {
|
| 665 |
+
"early_stopping_patience_counter": 1
|
| 666 |
+
}
|
| 667 |
+
},
|
| 668 |
+
"TrainerControl": {
|
| 669 |
+
"args": {
|
| 670 |
+
"should_epoch_stop": false,
|
| 671 |
+
"should_evaluate": false,
|
| 672 |
+
"should_log": false,
|
| 673 |
+
"should_save": true,
|
| 674 |
+
"should_training_stop": false
|
| 675 |
+
},
|
| 676 |
+
"attributes": {}
|
| 677 |
+
}
|
| 678 |
+
},
|
| 679 |
+
"total_flos": 1.6722992431104e+16,
|
| 680 |
+
"train_batch_size": 16,
|
| 681 |
+
"trial_name": null,
|
| 682 |
+
"trial_params": null
|
| 683 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e47fd9a569ce220c806b4a52069072c804efe830882485e57831413d7df87da6
|
| 3 |
+
size 5304
|
checkpoint-2000/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2500/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForQuestionAnswering"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"language": "english",
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"name": "Bert",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-2500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18c0d4ccf9777d8019baf7c748399dc7687ee02492e0440984e56f1b314d6669
|
| 3 |
+
size 435596088
|
checkpoint-2500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6da4bb3583c5236d187397ed63c30846b803000bb4e93712f1c57dafe9e4913b
|
| 3 |
+
size 871311930
|
checkpoint-2500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aac6ce6758a41be9c411ebc19f7bc46befaf0c549f6a057a6116e3dce102cc27
|
| 3 |
+
size 14244
|
checkpoint-2500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7ab928fb6fc03d23ed0a52a122112f8e7b9f1b5afe619387db540b707cec3ec
|
| 3 |
+
size 988
|
checkpoint-2500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bef47faf2118a55e23cfadff7c96ed79de25677acf2a06fc651b5ea4d808eb7
|
| 3 |
+
size 1064
|
checkpoint-2500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-2500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-2500/trainer_state.json
ADDED
|
@@ -0,0 +1,843 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2400,
|
| 3 |
+
"best_metric": 83.08410801910097,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.6622516556291391,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 2500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013245033112582781,
|
| 14 |
+
"grad_norm": 7.202547550201416,
|
| 15 |
+
"learning_rate": 9.998701986754968e-06,
|
| 16 |
+
"loss": 0.7094,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.026490066225165563,
|
| 21 |
+
"grad_norm": 9.662558555603027,
|
| 22 |
+
"learning_rate": 9.99737748344371e-06,
|
| 23 |
+
"loss": 0.68,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.026490066225165563,
|
| 28 |
+
"eval_HasAns_exact": 75.14792899408285,
|
| 29 |
+
"eval_HasAns_f1": 81.01957227969451,
|
| 30 |
+
"eval_HasAns_total": 15041,
|
| 31 |
+
"eval_best_exact": 75.14792899408285,
|
| 32 |
+
"eval_best_exact_thresh": 0.0,
|
| 33 |
+
"eval_best_f1": 81.01957227969451,
|
| 34 |
+
"eval_best_f1_thresh": 0.0,
|
| 35 |
+
"eval_exact": 75.14792899408285,
|
| 36 |
+
"eval_f1": 81.01957227969451,
|
| 37 |
+
"eval_loss": 0.5824803113937378,
|
| 38 |
+
"eval_runtime": 120.1578,
|
| 39 |
+
"eval_samples_per_second": 125.177,
|
| 40 |
+
"eval_steps_per_second": 7.831,
|
| 41 |
+
"eval_total": 15041,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.039735099337748346,
|
| 46 |
+
"grad_norm": 7.527211666107178,
|
| 47 |
+
"learning_rate": 9.996052980132451e-06,
|
| 48 |
+
"loss": 0.586,
|
| 49 |
+
"step": 150
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.052980132450331126,
|
| 53 |
+
"grad_norm": 5.419323921203613,
|
| 54 |
+
"learning_rate": 9.994728476821193e-06,
|
| 55 |
+
"loss": 0.6035,
|
| 56 |
+
"step": 200
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.052980132450331126,
|
| 60 |
+
"eval_HasAns_exact": 75.56013562927997,
|
| 61 |
+
"eval_HasAns_f1": 81.46153699536843,
|
| 62 |
+
"eval_HasAns_total": 15041,
|
| 63 |
+
"eval_best_exact": 75.56013562927997,
|
| 64 |
+
"eval_best_exact_thresh": 0.0,
|
| 65 |
+
"eval_best_f1": 81.46153699536843,
|
| 66 |
+
"eval_best_f1_thresh": 0.0,
|
| 67 |
+
"eval_exact": 75.56013562927997,
|
| 68 |
+
"eval_f1": 81.46153699536843,
|
| 69 |
+
"eval_loss": 0.5453085899353027,
|
| 70 |
+
"eval_runtime": 118.3719,
|
| 71 |
+
"eval_samples_per_second": 127.066,
|
| 72 |
+
"eval_steps_per_second": 7.95,
|
| 73 |
+
"eval_total": 15041,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.06622516556291391,
|
| 78 |
+
"grad_norm": 9.9099760055542,
|
| 79 |
+
"learning_rate": 9.993403973509935e-06,
|
| 80 |
+
"loss": 0.5616,
|
| 81 |
+
"step": 250
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.07947019867549669,
|
| 85 |
+
"grad_norm": 3.5184719562530518,
|
| 86 |
+
"learning_rate": 9.992079470198677e-06,
|
| 87 |
+
"loss": 0.5286,
|
| 88 |
+
"step": 300
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.07947019867549669,
|
| 92 |
+
"eval_HasAns_exact": 75.8526693703876,
|
| 93 |
+
"eval_HasAns_f1": 81.45745729278944,
|
| 94 |
+
"eval_HasAns_total": 15041,
|
| 95 |
+
"eval_best_exact": 75.8526693703876,
|
| 96 |
+
"eval_best_exact_thresh": 0.0,
|
| 97 |
+
"eval_best_f1": 81.45745729278944,
|
| 98 |
+
"eval_best_f1_thresh": 0.0,
|
| 99 |
+
"eval_exact": 75.8526693703876,
|
| 100 |
+
"eval_f1": 81.45745729278944,
|
| 101 |
+
"eval_loss": 0.5200360417366028,
|
| 102 |
+
"eval_runtime": 118.7441,
|
| 103 |
+
"eval_samples_per_second": 126.667,
|
| 104 |
+
"eval_steps_per_second": 7.925,
|
| 105 |
+
"eval_total": 15041,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.09271523178807947,
|
| 110 |
+
"grad_norm": 4.889588832855225,
|
| 111 |
+
"learning_rate": 9.990754966887418e-06,
|
| 112 |
+
"loss": 0.4943,
|
| 113 |
+
"step": 350
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.10596026490066225,
|
| 117 |
+
"grad_norm": 6.106881141662598,
|
| 118 |
+
"learning_rate": 9.98943046357616e-06,
|
| 119 |
+
"loss": 0.5957,
|
| 120 |
+
"step": 400
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.10596026490066225,
|
| 124 |
+
"eval_HasAns_exact": 76.39784588790639,
|
| 125 |
+
"eval_HasAns_f1": 81.91199737367272,
|
| 126 |
+
"eval_HasAns_total": 15041,
|
| 127 |
+
"eval_best_exact": 76.39784588790639,
|
| 128 |
+
"eval_best_exact_thresh": 0.0,
|
| 129 |
+
"eval_best_f1": 81.91199737367272,
|
| 130 |
+
"eval_best_f1_thresh": 0.0,
|
| 131 |
+
"eval_exact": 76.39784588790639,
|
| 132 |
+
"eval_f1": 81.91199737367272,
|
| 133 |
+
"eval_loss": 0.49782633781433105,
|
| 134 |
+
"eval_runtime": 118.2655,
|
| 135 |
+
"eval_samples_per_second": 127.18,
|
| 136 |
+
"eval_steps_per_second": 7.957,
|
| 137 |
+
"eval_total": 15041,
|
| 138 |
+
"step": 400
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.11920529801324503,
|
| 142 |
+
"grad_norm": 6.963775634765625,
|
| 143 |
+
"learning_rate": 9.988105960264902e-06,
|
| 144 |
+
"loss": 0.5675,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.13245033112582782,
|
| 149 |
+
"grad_norm": 7.559121608734131,
|
| 150 |
+
"learning_rate": 9.986781456953643e-06,
|
| 151 |
+
"loss": 0.5173,
|
| 152 |
+
"step": 500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.13245033112582782,
|
| 156 |
+
"eval_HasAns_exact": 75.94574828801277,
|
| 157 |
+
"eval_HasAns_f1": 81.43615773338746,
|
| 158 |
+
"eval_HasAns_total": 15041,
|
| 159 |
+
"eval_best_exact": 75.94574828801277,
|
| 160 |
+
"eval_best_exact_thresh": 0.0,
|
| 161 |
+
"eval_best_f1": 81.43615773338746,
|
| 162 |
+
"eval_best_f1_thresh": 0.0,
|
| 163 |
+
"eval_exact": 75.94574828801277,
|
| 164 |
+
"eval_f1": 81.43615773338746,
|
| 165 |
+
"eval_loss": 0.4918788969516754,
|
| 166 |
+
"eval_runtime": 118.4997,
|
| 167 |
+
"eval_samples_per_second": 126.929,
|
| 168 |
+
"eval_steps_per_second": 7.941,
|
| 169 |
+
"eval_total": 15041,
|
| 170 |
+
"step": 500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1456953642384106,
|
| 174 |
+
"grad_norm": 5.2273268699646,
|
| 175 |
+
"learning_rate": 9.985456953642385e-06,
|
| 176 |
+
"loss": 0.5059,
|
| 177 |
+
"step": 550
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.15894039735099338,
|
| 181 |
+
"grad_norm": 13.281010627746582,
|
| 182 |
+
"learning_rate": 9.984132450331127e-06,
|
| 183 |
+
"loss": 0.4847,
|
| 184 |
+
"step": 600
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.15894039735099338,
|
| 188 |
+
"eval_HasAns_exact": 76.42443986437073,
|
| 189 |
+
"eval_HasAns_f1": 81.98556495448281,
|
| 190 |
+
"eval_HasAns_total": 15041,
|
| 191 |
+
"eval_best_exact": 76.42443986437073,
|
| 192 |
+
"eval_best_exact_thresh": 0.0,
|
| 193 |
+
"eval_best_f1": 81.98556495448281,
|
| 194 |
+
"eval_best_f1_thresh": 0.0,
|
| 195 |
+
"eval_exact": 76.42443986437073,
|
| 196 |
+
"eval_f1": 81.98556495448281,
|
| 197 |
+
"eval_loss": 0.48446840047836304,
|
| 198 |
+
"eval_runtime": 118.3928,
|
| 199 |
+
"eval_samples_per_second": 127.043,
|
| 200 |
+
"eval_steps_per_second": 7.948,
|
| 201 |
+
"eval_total": 15041,
|
| 202 |
+
"step": 600
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.17218543046357615,
|
| 206 |
+
"grad_norm": 18.346189498901367,
|
| 207 |
+
"learning_rate": 9.982807947019869e-06,
|
| 208 |
+
"loss": 0.4857,
|
| 209 |
+
"step": 650
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.18543046357615894,
|
| 213 |
+
"grad_norm": 11.825461387634277,
|
| 214 |
+
"learning_rate": 9.98148344370861e-06,
|
| 215 |
+
"loss": 0.4764,
|
| 216 |
+
"step": 700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.18543046357615894,
|
| 220 |
+
"eval_HasAns_exact": 76.54411275846022,
|
| 221 |
+
"eval_HasAns_f1": 82.14175205695277,
|
| 222 |
+
"eval_HasAns_total": 15041,
|
| 223 |
+
"eval_best_exact": 76.54411275846022,
|
| 224 |
+
"eval_best_exact_thresh": 0.0,
|
| 225 |
+
"eval_best_f1": 82.14175205695277,
|
| 226 |
+
"eval_best_f1_thresh": 0.0,
|
| 227 |
+
"eval_exact": 76.54411275846022,
|
| 228 |
+
"eval_f1": 82.14175205695277,
|
| 229 |
+
"eval_loss": 0.4717336595058441,
|
| 230 |
+
"eval_runtime": 118.5022,
|
| 231 |
+
"eval_samples_per_second": 126.926,
|
| 232 |
+
"eval_steps_per_second": 7.941,
|
| 233 |
+
"eval_total": 15041,
|
| 234 |
+
"step": 700
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1986754966887417,
|
| 238 |
+
"grad_norm": 7.6593475341796875,
|
| 239 |
+
"learning_rate": 9.980158940397352e-06,
|
| 240 |
+
"loss": 0.486,
|
| 241 |
+
"step": 750
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2119205298013245,
|
| 245 |
+
"grad_norm": 6.091643333435059,
|
| 246 |
+
"learning_rate": 9.978834437086094e-06,
|
| 247 |
+
"loss": 0.4707,
|
| 248 |
+
"step": 800
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2119205298013245,
|
| 252 |
+
"eval_HasAns_exact": 76.86324047603217,
|
| 253 |
+
"eval_HasAns_f1": 82.2239211673303,
|
| 254 |
+
"eval_HasAns_total": 15041,
|
| 255 |
+
"eval_best_exact": 76.86324047603217,
|
| 256 |
+
"eval_best_exact_thresh": 0.0,
|
| 257 |
+
"eval_best_f1": 82.2239211673303,
|
| 258 |
+
"eval_best_f1_thresh": 0.0,
|
| 259 |
+
"eval_exact": 76.86324047603217,
|
| 260 |
+
"eval_f1": 82.2239211673303,
|
| 261 |
+
"eval_loss": 0.4601329565048218,
|
| 262 |
+
"eval_runtime": 119.1574,
|
| 263 |
+
"eval_samples_per_second": 126.228,
|
| 264 |
+
"eval_steps_per_second": 7.897,
|
| 265 |
+
"eval_total": 15041,
|
| 266 |
+
"step": 800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2251655629139073,
|
| 270 |
+
"grad_norm": 4.863158702850342,
|
| 271 |
+
"learning_rate": 9.977509933774835e-06,
|
| 272 |
+
"loss": 0.5042,
|
| 273 |
+
"step": 850
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.23841059602649006,
|
| 277 |
+
"grad_norm": 5.747822284698486,
|
| 278 |
+
"learning_rate": 9.976185430463577e-06,
|
| 279 |
+
"loss": 0.4725,
|
| 280 |
+
"step": 900
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.23841059602649006,
|
| 284 |
+
"eval_HasAns_exact": 76.98291337012166,
|
| 285 |
+
"eval_HasAns_f1": 82.39127215176644,
|
| 286 |
+
"eval_HasAns_total": 15041,
|
| 287 |
+
"eval_best_exact": 76.98291337012166,
|
| 288 |
+
"eval_best_exact_thresh": 0.0,
|
| 289 |
+
"eval_best_f1": 82.39127215176644,
|
| 290 |
+
"eval_best_f1_thresh": 0.0,
|
| 291 |
+
"eval_exact": 76.98291337012166,
|
| 292 |
+
"eval_f1": 82.39127215176644,
|
| 293 |
+
"eval_loss": 0.45768168568611145,
|
| 294 |
+
"eval_runtime": 118.7264,
|
| 295 |
+
"eval_samples_per_second": 126.686,
|
| 296 |
+
"eval_steps_per_second": 7.926,
|
| 297 |
+
"eval_total": 15041,
|
| 298 |
+
"step": 900
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.25165562913907286,
|
| 302 |
+
"grad_norm": 14.890397071838379,
|
| 303 |
+
"learning_rate": 9.974860927152319e-06,
|
| 304 |
+
"loss": 0.4476,
|
| 305 |
+
"step": 950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.26490066225165565,
|
| 309 |
+
"grad_norm": 8.837732315063477,
|
| 310 |
+
"learning_rate": 9.97353642384106e-06,
|
| 311 |
+
"loss": 0.4905,
|
| 312 |
+
"step": 1000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.26490066225165565,
|
| 316 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 317 |
+
"eval_HasAns_f1": 82.37799175072134,
|
| 318 |
+
"eval_HasAns_total": 15041,
|
| 319 |
+
"eval_best_exact": 77.10923475832723,
|
| 320 |
+
"eval_best_exact_thresh": 0.0,
|
| 321 |
+
"eval_best_f1": 82.37799175072134,
|
| 322 |
+
"eval_best_f1_thresh": 0.0,
|
| 323 |
+
"eval_exact": 77.10923475832723,
|
| 324 |
+
"eval_f1": 82.37799175072134,
|
| 325 |
+
"eval_loss": 0.44477367401123047,
|
| 326 |
+
"eval_runtime": 118.302,
|
| 327 |
+
"eval_samples_per_second": 127.141,
|
| 328 |
+
"eval_steps_per_second": 7.954,
|
| 329 |
+
"eval_total": 15041,
|
| 330 |
+
"step": 1000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.2781456953642384,
|
| 334 |
+
"grad_norm": 5.750672817230225,
|
| 335 |
+
"learning_rate": 9.972211920529802e-06,
|
| 336 |
+
"loss": 0.5041,
|
| 337 |
+
"step": 1050
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.2913907284768212,
|
| 341 |
+
"grad_norm": 5.720890998840332,
|
| 342 |
+
"learning_rate": 9.970887417218544e-06,
|
| 343 |
+
"loss": 0.5156,
|
| 344 |
+
"step": 1100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.2913907284768212,
|
| 348 |
+
"eval_HasAns_exact": 77.0826407818629,
|
| 349 |
+
"eval_HasAns_f1": 82.36916366838726,
|
| 350 |
+
"eval_HasAns_total": 15041,
|
| 351 |
+
"eval_best_exact": 77.0826407818629,
|
| 352 |
+
"eval_best_exact_thresh": 0.0,
|
| 353 |
+
"eval_best_f1": 82.36916366838726,
|
| 354 |
+
"eval_best_f1_thresh": 0.0,
|
| 355 |
+
"eval_exact": 77.0826407818629,
|
| 356 |
+
"eval_f1": 82.36916366838726,
|
| 357 |
+
"eval_loss": 0.43873801827430725,
|
| 358 |
+
"eval_runtime": 118.2856,
|
| 359 |
+
"eval_samples_per_second": 127.158,
|
| 360 |
+
"eval_steps_per_second": 7.955,
|
| 361 |
+
"eval_total": 15041,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.304635761589404,
|
| 366 |
+
"grad_norm": 6.990389823913574,
|
| 367 |
+
"learning_rate": 9.969562913907286e-06,
|
| 368 |
+
"loss": 0.4507,
|
| 369 |
+
"step": 1150
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.31788079470198677,
|
| 373 |
+
"grad_norm": 11.490405082702637,
|
| 374 |
+
"learning_rate": 9.968238410596027e-06,
|
| 375 |
+
"loss": 0.4482,
|
| 376 |
+
"step": 1200
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.31788079470198677,
|
| 380 |
+
"eval_HasAns_exact": 77.01615584070208,
|
| 381 |
+
"eval_HasAns_f1": 82.29906445927755,
|
| 382 |
+
"eval_HasAns_total": 15041,
|
| 383 |
+
"eval_best_exact": 77.01615584070208,
|
| 384 |
+
"eval_best_exact_thresh": 0.0,
|
| 385 |
+
"eval_best_f1": 82.29906445927755,
|
| 386 |
+
"eval_best_f1_thresh": 0.0,
|
| 387 |
+
"eval_exact": 77.01615584070208,
|
| 388 |
+
"eval_f1": 82.29906445927755,
|
| 389 |
+
"eval_loss": 0.42996492981910706,
|
| 390 |
+
"eval_runtime": 118.4966,
|
| 391 |
+
"eval_samples_per_second": 126.932,
|
| 392 |
+
"eval_steps_per_second": 7.941,
|
| 393 |
+
"eval_total": 15041,
|
| 394 |
+
"step": 1200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.33112582781456956,
|
| 398 |
+
"grad_norm": 2.8688676357269287,
|
| 399 |
+
"learning_rate": 9.966913907284769e-06,
|
| 400 |
+
"loss": 0.4593,
|
| 401 |
+
"step": 1250
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3443708609271523,
|
| 405 |
+
"grad_norm": 7.523802280426025,
|
| 406 |
+
"learning_rate": 9.96558940397351e-06,
|
| 407 |
+
"loss": 0.4548,
|
| 408 |
+
"step": 1300
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3443708609271523,
|
| 412 |
+
"eval_HasAns_exact": 76.94967089954126,
|
| 413 |
+
"eval_HasAns_f1": 82.17470826997452,
|
| 414 |
+
"eval_HasAns_total": 15041,
|
| 415 |
+
"eval_best_exact": 76.94967089954126,
|
| 416 |
+
"eval_best_exact_thresh": 0.0,
|
| 417 |
+
"eval_best_f1": 82.17470826997452,
|
| 418 |
+
"eval_best_f1_thresh": 0.0,
|
| 419 |
+
"eval_exact": 76.94967089954126,
|
| 420 |
+
"eval_f1": 82.17470826997452,
|
| 421 |
+
"eval_loss": 0.4349832832813263,
|
| 422 |
+
"eval_runtime": 118.6592,
|
| 423 |
+
"eval_samples_per_second": 126.758,
|
| 424 |
+
"eval_steps_per_second": 7.93,
|
| 425 |
+
"eval_total": 15041,
|
| 426 |
+
"step": 1300
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.3576158940397351,
|
| 430 |
+
"grad_norm": 10.102520942687988,
|
| 431 |
+
"learning_rate": 9.964264900662252e-06,
|
| 432 |
+
"loss": 0.4935,
|
| 433 |
+
"step": 1350
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.3708609271523179,
|
| 437 |
+
"grad_norm": 8.859588623046875,
|
| 438 |
+
"learning_rate": 9.962940397350994e-06,
|
| 439 |
+
"loss": 0.4344,
|
| 440 |
+
"step": 1400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.3708609271523179,
|
| 444 |
+
"eval_HasAns_exact": 77.19566518183632,
|
| 445 |
+
"eval_HasAns_f1": 82.55212784144531,
|
| 446 |
+
"eval_HasAns_total": 15041,
|
| 447 |
+
"eval_best_exact": 77.19566518183632,
|
| 448 |
+
"eval_best_exact_thresh": 0.0,
|
| 449 |
+
"eval_best_f1": 82.55212784144531,
|
| 450 |
+
"eval_best_f1_thresh": 0.0,
|
| 451 |
+
"eval_exact": 77.19566518183632,
|
| 452 |
+
"eval_f1": 82.55212784144531,
|
| 453 |
+
"eval_loss": 0.42366155982017517,
|
| 454 |
+
"eval_runtime": 119.1071,
|
| 455 |
+
"eval_samples_per_second": 126.281,
|
| 456 |
+
"eval_steps_per_second": 7.9,
|
| 457 |
+
"eval_total": 15041,
|
| 458 |
+
"step": 1400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3841059602649007,
|
| 462 |
+
"grad_norm": 5.980023384094238,
|
| 463 |
+
"learning_rate": 9.961615894039736e-06,
|
| 464 |
+
"loss": 0.4936,
|
| 465 |
+
"step": 1450
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3973509933774834,
|
| 469 |
+
"grad_norm": 5.025954246520996,
|
| 470 |
+
"learning_rate": 9.960291390728478e-06,
|
| 471 |
+
"loss": 0.4431,
|
| 472 |
+
"step": 1500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.3973509933774834,
|
| 476 |
+
"eval_HasAns_exact": 77.10923475832723,
|
| 477 |
+
"eval_HasAns_f1": 82.41666220398032,
|
| 478 |
+
"eval_HasAns_total": 15041,
|
| 479 |
+
"eval_best_exact": 77.10923475832723,
|
| 480 |
+
"eval_best_exact_thresh": 0.0,
|
| 481 |
+
"eval_best_f1": 82.41666220398032,
|
| 482 |
+
"eval_best_f1_thresh": 0.0,
|
| 483 |
+
"eval_exact": 77.10923475832723,
|
| 484 |
+
"eval_f1": 82.41666220398032,
|
| 485 |
+
"eval_loss": 0.4237441122531891,
|
| 486 |
+
"eval_runtime": 118.7094,
|
| 487 |
+
"eval_samples_per_second": 126.704,
|
| 488 |
+
"eval_steps_per_second": 7.927,
|
| 489 |
+
"eval_total": 15041,
|
| 490 |
+
"step": 1500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.4105960264900662,
|
| 494 |
+
"grad_norm": 13.039314270019531,
|
| 495 |
+
"learning_rate": 9.95896688741722e-06,
|
| 496 |
+
"loss": 0.4907,
|
| 497 |
+
"step": 1550
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.423841059602649,
|
| 501 |
+
"grad_norm": 8.910651206970215,
|
| 502 |
+
"learning_rate": 9.957642384105961e-06,
|
| 503 |
+
"loss": 0.4462,
|
| 504 |
+
"step": 1600
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.423841059602649,
|
| 508 |
+
"eval_HasAns_exact": 77.27544711122931,
|
| 509 |
+
"eval_HasAns_f1": 82.45441363866084,
|
| 510 |
+
"eval_HasAns_total": 15041,
|
| 511 |
+
"eval_best_exact": 77.27544711122931,
|
| 512 |
+
"eval_best_exact_thresh": 0.0,
|
| 513 |
+
"eval_best_f1": 82.45441363866084,
|
| 514 |
+
"eval_best_f1_thresh": 0.0,
|
| 515 |
+
"eval_exact": 77.27544711122931,
|
| 516 |
+
"eval_f1": 82.45441363866084,
|
| 517 |
+
"eval_loss": 0.4120015501976013,
|
| 518 |
+
"eval_runtime": 118.2118,
|
| 519 |
+
"eval_samples_per_second": 127.238,
|
| 520 |
+
"eval_steps_per_second": 7.96,
|
| 521 |
+
"eval_total": 15041,
|
| 522 |
+
"step": 1600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.4370860927152318,
|
| 526 |
+
"grad_norm": 7.536431312561035,
|
| 527 |
+
"learning_rate": 9.956317880794703e-06,
|
| 528 |
+
"loss": 0.423,
|
| 529 |
+
"step": 1650
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.4503311258278146,
|
| 533 |
+
"grad_norm": 6.14304780960083,
|
| 534 |
+
"learning_rate": 9.954993377483444e-06,
|
| 535 |
+
"loss": 0.4353,
|
| 536 |
+
"step": 1700
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.4503311258278146,
|
| 540 |
+
"eval_HasAns_exact": 77.64111428761386,
|
| 541 |
+
"eval_HasAns_f1": 82.82953930481963,
|
| 542 |
+
"eval_HasAns_total": 15041,
|
| 543 |
+
"eval_best_exact": 77.64111428761386,
|
| 544 |
+
"eval_best_exact_thresh": 0.0,
|
| 545 |
+
"eval_best_f1": 82.82953930481963,
|
| 546 |
+
"eval_best_f1_thresh": 0.0,
|
| 547 |
+
"eval_exact": 77.64111428761386,
|
| 548 |
+
"eval_f1": 82.82953930481963,
|
| 549 |
+
"eval_loss": 0.4078502357006073,
|
| 550 |
+
"eval_runtime": 118.4213,
|
| 551 |
+
"eval_samples_per_second": 127.013,
|
| 552 |
+
"eval_steps_per_second": 7.946,
|
| 553 |
+
"eval_total": 15041,
|
| 554 |
+
"step": 1700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.46357615894039733,
|
| 558 |
+
"grad_norm": 7.738461017608643,
|
| 559 |
+
"learning_rate": 9.953668874172186e-06,
|
| 560 |
+
"loss": 0.4166,
|
| 561 |
+
"step": 1750
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.4768211920529801,
|
| 565 |
+
"grad_norm": 5.486303329467773,
|
| 566 |
+
"learning_rate": 9.952344370860928e-06,
|
| 567 |
+
"loss": 0.4344,
|
| 568 |
+
"step": 1800
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.4768211920529801,
|
| 572 |
+
"eval_HasAns_exact": 77.53473838175653,
|
| 573 |
+
"eval_HasAns_f1": 82.83341057132324,
|
| 574 |
+
"eval_HasAns_total": 15041,
|
| 575 |
+
"eval_best_exact": 77.53473838175653,
|
| 576 |
+
"eval_best_exact_thresh": 0.0,
|
| 577 |
+
"eval_best_f1": 82.83341057132324,
|
| 578 |
+
"eval_best_f1_thresh": 0.0,
|
| 579 |
+
"eval_exact": 77.53473838175653,
|
| 580 |
+
"eval_f1": 82.83341057132324,
|
| 581 |
+
"eval_loss": 0.4005696475505829,
|
| 582 |
+
"eval_runtime": 118.4606,
|
| 583 |
+
"eval_samples_per_second": 126.97,
|
| 584 |
+
"eval_steps_per_second": 7.944,
|
| 585 |
+
"eval_total": 15041,
|
| 586 |
+
"step": 1800
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4900662251655629,
|
| 590 |
+
"grad_norm": 11.653322219848633,
|
| 591 |
+
"learning_rate": 9.95101986754967e-06,
|
| 592 |
+
"loss": 0.444,
|
| 593 |
+
"step": 1850
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.5033112582781457,
|
| 597 |
+
"grad_norm": 9.203779220581055,
|
| 598 |
+
"learning_rate": 9.949695364238411e-06,
|
| 599 |
+
"loss": 0.3835,
|
| 600 |
+
"step": 1900
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.5033112582781457,
|
| 604 |
+
"eval_HasAns_exact": 77.813975134632,
|
| 605 |
+
"eval_HasAns_f1": 82.9212903481507,
|
| 606 |
+
"eval_HasAns_total": 15041,
|
| 607 |
+
"eval_best_exact": 77.813975134632,
|
| 608 |
+
"eval_best_exact_thresh": 0.0,
|
| 609 |
+
"eval_best_f1": 82.9212903481507,
|
| 610 |
+
"eval_best_f1_thresh": 0.0,
|
| 611 |
+
"eval_exact": 77.813975134632,
|
| 612 |
+
"eval_f1": 82.9212903481507,
|
| 613 |
+
"eval_loss": 0.40120014548301697,
|
| 614 |
+
"eval_runtime": 118.8575,
|
| 615 |
+
"eval_samples_per_second": 126.547,
|
| 616 |
+
"eval_steps_per_second": 7.917,
|
| 617 |
+
"eval_total": 15041,
|
| 618 |
+
"step": 1900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5165562913907285,
|
| 622 |
+
"grad_norm": 11.891190528869629,
|
| 623 |
+
"learning_rate": 9.948370860927153e-06,
|
| 624 |
+
"loss": 0.4682,
|
| 625 |
+
"step": 1950
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5298013245033113,
|
| 629 |
+
"grad_norm": 7.65544319152832,
|
| 630 |
+
"learning_rate": 9.947046357615895e-06,
|
| 631 |
+
"loss": 0.4618,
|
| 632 |
+
"step": 2000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5298013245033113,
|
| 636 |
+
"eval_HasAns_exact": 77.63446579349777,
|
| 637 |
+
"eval_HasAns_f1": 82.84673293378647,
|
| 638 |
+
"eval_HasAns_total": 15041,
|
| 639 |
+
"eval_best_exact": 77.63446579349777,
|
| 640 |
+
"eval_best_exact_thresh": 0.0,
|
| 641 |
+
"eval_best_f1": 82.84673293378647,
|
| 642 |
+
"eval_best_f1_thresh": 0.0,
|
| 643 |
+
"eval_exact": 77.63446579349777,
|
| 644 |
+
"eval_f1": 82.84673293378647,
|
| 645 |
+
"eval_loss": 0.38913631439208984,
|
| 646 |
+
"eval_runtime": 118.6931,
|
| 647 |
+
"eval_samples_per_second": 126.722,
|
| 648 |
+
"eval_steps_per_second": 7.928,
|
| 649 |
+
"eval_total": 15041,
|
| 650 |
+
"step": 2000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.543046357615894,
|
| 654 |
+
"grad_norm": 8.86410140991211,
|
| 655 |
+
"learning_rate": 9.945721854304636e-06,
|
| 656 |
+
"loss": 0.4094,
|
| 657 |
+
"step": 2050
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.5562913907284768,
|
| 661 |
+
"grad_norm": 9.725333213806152,
|
| 662 |
+
"learning_rate": 9.944423841059604e-06,
|
| 663 |
+
"loss": 0.4156,
|
| 664 |
+
"step": 2100
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.5562913907284768,
|
| 668 |
+
"eval_HasAns_exact": 77.56133235822087,
|
| 669 |
+
"eval_HasAns_f1": 82.70805592907325,
|
| 670 |
+
"eval_HasAns_total": 15041,
|
| 671 |
+
"eval_best_exact": 77.56133235822087,
|
| 672 |
+
"eval_best_exact_thresh": 0.0,
|
| 673 |
+
"eval_best_f1": 82.70805592907325,
|
| 674 |
+
"eval_best_f1_thresh": 0.0,
|
| 675 |
+
"eval_exact": 77.56133235822087,
|
| 676 |
+
"eval_f1": 82.70805592907325,
|
| 677 |
+
"eval_loss": 0.38439399003982544,
|
| 678 |
+
"eval_runtime": 118.8277,
|
| 679 |
+
"eval_samples_per_second": 126.578,
|
| 680 |
+
"eval_steps_per_second": 7.919,
|
| 681 |
+
"eval_total": 15041,
|
| 682 |
+
"step": 2100
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5695364238410596,
|
| 686 |
+
"grad_norm": 3.56200909614563,
|
| 687 |
+
"learning_rate": 9.943099337748345e-06,
|
| 688 |
+
"loss": 0.3851,
|
| 689 |
+
"step": 2150
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5827814569536424,
|
| 693 |
+
"grad_norm": 5.532958984375,
|
| 694 |
+
"learning_rate": 9.941774834437087e-06,
|
| 695 |
+
"loss": 0.4051,
|
| 696 |
+
"step": 2200
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5827814569536424,
|
| 700 |
+
"eval_HasAns_exact": 77.88710856990892,
|
| 701 |
+
"eval_HasAns_f1": 82.85876664892294,
|
| 702 |
+
"eval_HasAns_total": 15041,
|
| 703 |
+
"eval_best_exact": 77.88710856990892,
|
| 704 |
+
"eval_best_exact_thresh": 0.0,
|
| 705 |
+
"eval_best_f1": 82.85876664892294,
|
| 706 |
+
"eval_best_f1_thresh": 0.0,
|
| 707 |
+
"eval_exact": 77.88710856990892,
|
| 708 |
+
"eval_f1": 82.85876664892294,
|
| 709 |
+
"eval_loss": 0.3852081000804901,
|
| 710 |
+
"eval_runtime": 118.3671,
|
| 711 |
+
"eval_samples_per_second": 127.071,
|
| 712 |
+
"eval_steps_per_second": 7.95,
|
| 713 |
+
"eval_total": 15041,
|
| 714 |
+
"step": 2200
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.5960264900662252,
|
| 718 |
+
"grad_norm": 6.847942352294922,
|
| 719 |
+
"learning_rate": 9.940450331125829e-06,
|
| 720 |
+
"loss": 0.375,
|
| 721 |
+
"step": 2250
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.609271523178808,
|
| 725 |
+
"grad_norm": 9.866053581237793,
|
| 726 |
+
"learning_rate": 9.93912582781457e-06,
|
| 727 |
+
"loss": 0.4071,
|
| 728 |
+
"step": 2300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.609271523178808,
|
| 732 |
+
"eval_HasAns_exact": 77.68100525231036,
|
| 733 |
+
"eval_HasAns_f1": 82.83004710532218,
|
| 734 |
+
"eval_HasAns_total": 15041,
|
| 735 |
+
"eval_best_exact": 77.68100525231036,
|
| 736 |
+
"eval_best_exact_thresh": 0.0,
|
| 737 |
+
"eval_best_f1": 82.83004710532218,
|
| 738 |
+
"eval_best_f1_thresh": 0.0,
|
| 739 |
+
"eval_exact": 77.68100525231036,
|
| 740 |
+
"eval_f1": 82.83004710532218,
|
| 741 |
+
"eval_loss": 0.38325873017311096,
|
| 742 |
+
"eval_runtime": 118.246,
|
| 743 |
+
"eval_samples_per_second": 127.201,
|
| 744 |
+
"eval_steps_per_second": 7.958,
|
| 745 |
+
"eval_total": 15041,
|
| 746 |
+
"step": 2300
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.6225165562913907,
|
| 750 |
+
"grad_norm": 10.307299613952637,
|
| 751 |
+
"learning_rate": 9.937801324503312e-06,
|
| 752 |
+
"loss": 0.4241,
|
| 753 |
+
"step": 2350
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.6357615894039735,
|
| 757 |
+
"grad_norm": 4.856252670288086,
|
| 758 |
+
"learning_rate": 9.936476821192054e-06,
|
| 759 |
+
"loss": 0.3738,
|
| 760 |
+
"step": 2400
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.6357615894039735,
|
| 764 |
+
"eval_HasAns_exact": 77.893757064025,
|
| 765 |
+
"eval_HasAns_f1": 83.08410801910097,
|
| 766 |
+
"eval_HasAns_total": 15041,
|
| 767 |
+
"eval_best_exact": 77.893757064025,
|
| 768 |
+
"eval_best_exact_thresh": 0.0,
|
| 769 |
+
"eval_best_f1": 83.08410801910097,
|
| 770 |
+
"eval_best_f1_thresh": 0.0,
|
| 771 |
+
"eval_exact": 77.893757064025,
|
| 772 |
+
"eval_f1": 83.08410801910097,
|
| 773 |
+
"eval_loss": 0.38139665126800537,
|
| 774 |
+
"eval_runtime": 118.3449,
|
| 775 |
+
"eval_samples_per_second": 127.095,
|
| 776 |
+
"eval_steps_per_second": 7.951,
|
| 777 |
+
"eval_total": 15041,
|
| 778 |
+
"step": 2400
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.6490066225165563,
|
| 782 |
+
"grad_norm": 9.251922607421875,
|
| 783 |
+
"learning_rate": 9.93517880794702e-06,
|
| 784 |
+
"loss": 0.3992,
|
| 785 |
+
"step": 2450
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.6622516556291391,
|
| 789 |
+
"grad_norm": 7.622641563415527,
|
| 790 |
+
"learning_rate": 9.933854304635763e-06,
|
| 791 |
+
"loss": 0.4027,
|
| 792 |
+
"step": 2500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.6622516556291391,
|
| 796 |
+
"eval_HasAns_exact": 77.60122332291736,
|
| 797 |
+
"eval_HasAns_f1": 82.61025686778665,
|
| 798 |
+
"eval_HasAns_total": 15041,
|
| 799 |
+
"eval_best_exact": 77.60122332291736,
|
| 800 |
+
"eval_best_exact_thresh": 0.0,
|
| 801 |
+
"eval_best_f1": 82.61025686778665,
|
| 802 |
+
"eval_best_f1_thresh": 0.0,
|
| 803 |
+
"eval_exact": 77.60122332291736,
|
| 804 |
+
"eval_f1": 82.61025686778665,
|
| 805 |
+
"eval_loss": 0.3716657757759094,
|
| 806 |
+
"eval_runtime": 119.038,
|
| 807 |
+
"eval_samples_per_second": 126.355,
|
| 808 |
+
"eval_steps_per_second": 7.905,
|
| 809 |
+
"eval_total": 15041,
|
| 810 |
+
"step": 2500
|
| 811 |
+
}
|
| 812 |
+
],
|
| 813 |
+
"logging_steps": 50,
|
| 814 |
+
"max_steps": 377500,
|
| 815 |
+
"num_input_tokens_seen": 0,
|
| 816 |
+
"num_train_epochs": 100,
|
| 817 |
+
"save_steps": 500,
|
| 818 |
+
"stateful_callbacks": {
|
| 819 |
+
"EarlyStoppingCallback": {
|
| 820 |
+
"args": {
|
| 821 |
+
"early_stopping_patience": 10,
|
| 822 |
+
"early_stopping_threshold": 0.0
|
| 823 |
+
},
|
| 824 |
+
"attributes": {
|
| 825 |
+
"early_stopping_patience_counter": 1
|
| 826 |
+
}
|
| 827 |
+
},
|
| 828 |
+
"TrainerControl": {
|
| 829 |
+
"args": {
|
| 830 |
+
"should_epoch_stop": false,
|
| 831 |
+
"should_evaluate": false,
|
| 832 |
+
"should_log": false,
|
| 833 |
+
"should_save": true,
|
| 834 |
+
"should_training_stop": false
|
| 835 |
+
},
|
| 836 |
+
"attributes": {}
|
| 837 |
+
}
|
| 838 |
+
},
|
| 839 |
+
"total_flos": 2.090374053888e+16,
|
| 840 |
+
"train_batch_size": 16,
|
| 841 |
+
"trial_name": null,
|
| 842 |
+
"trial_params": null
|
| 843 |
+
}
|
checkpoint-2500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e47fd9a569ce220c806b4a52069072c804efe830882485e57831413d7df87da6
|
| 3 |
+
size 5304
|
checkpoint-2500/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3000/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForQuestionAnswering"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"language": "english",
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"name": "Bert",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-3000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c50a565915ef163cb8b39237e23c92e89676d43d36f175d525ddda208a601a07
|
| 3 |
+
size 435596088
|