Add files using upload-large-folder tool
Browse files- checkpoint-800/config.json +39 -0
- checkpoint-800/model.safetensors +3 -0
- checkpoint-800/optimizer.pt +3 -0
- checkpoint-800/rng_state.pth +3 -0
- checkpoint-800/scaler.pt +3 -0
- checkpoint-800/scheduler.pt +3 -0
- checkpoint-800/trainer_state.json +210 -0
- checkpoint-800/training_args.bin +3 -0
- checkpoint-900/config.json +39 -0
- checkpoint-900/model.safetensors +3 -0
- checkpoint-900/optimizer.pt +3 -0
- checkpoint-900/rng_state.pth +3 -0
- checkpoint-900/scaler.pt +3 -0
- checkpoint-900/scheduler.pt +3 -0
- checkpoint-900/trainer_state.json +232 -0
- checkpoint-900/training_args.bin +3 -0
checkpoint-800/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"DebertaV2ForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": null,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"eos_token_id": null,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 4096,
|
| 14 |
+
"layer_norm_eps": 1e-07,
|
| 15 |
+
"legacy": true,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"max_relative_positions": -1,
|
| 18 |
+
"model_type": "deberta-v2",
|
| 19 |
+
"norm_rel_ebd": "layer_norm",
|
| 20 |
+
"num_attention_heads": 16,
|
| 21 |
+
"num_hidden_layers": 24,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"pooler_dropout": 0,
|
| 24 |
+
"pooler_hidden_act": "gelu",
|
| 25 |
+
"pooler_hidden_size": 1024,
|
| 26 |
+
"pos_att_type": [
|
| 27 |
+
"p2c",
|
| 28 |
+
"c2p"
|
| 29 |
+
],
|
| 30 |
+
"position_biased_input": false,
|
| 31 |
+
"position_buckets": 256,
|
| 32 |
+
"relative_attention": true,
|
| 33 |
+
"share_att_key": true,
|
| 34 |
+
"tie_word_embeddings": true,
|
| 35 |
+
"transformers_version": "5.12.1",
|
| 36 |
+
"type_vocab_size": 0,
|
| 37 |
+
"use_cache": false,
|
| 38 |
+
"vocab_size": 128100
|
| 39 |
+
}
|
checkpoint-800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ad7400d4c8df4b93cc65d0177a8dd3c355f400547e171236c9e2d768d3b7aa8
|
| 3 |
+
size 1740304440
|
checkpoint-800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3444619b0f82bb48bbb48c7a7b5e3e5eed332fb0f71c1a88029d1769f489d433
|
| 3 |
+
size 2271441914
|
checkpoint-800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd683dc2afe44c387434354fb9ef18e1cd16d5d4f98b059733fd29b443979aa8
|
| 3 |
+
size 14180
|
checkpoint-800/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:deae275023ea82834d851b29526bc8569e8b6f6f0c6b90510f704ca095deffa1
|
| 3 |
+
size 988
|
checkpoint-800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ab023a6a7fdadf2f6eaf4d2fcedabe8cf9c521aac94112f81e805c4ad1f6701
|
| 3 |
+
size 1064
|
checkpoint-800/trainer_state.json
ADDED
|
@@ -0,0 +1,210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.512,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 25.09992027282715,
|
| 15 |
+
"learning_rate": 4.3999999999999997e-07,
|
| 16 |
+
"loss": 1.0546456146240235,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.064,
|
| 21 |
+
"grad_norm": 44.00643539428711,
|
| 22 |
+
"learning_rate": 9.399999999999999e-07,
|
| 23 |
+
"loss": 1.2045201873779297,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.064,
|
| 28 |
+
"eval_loss": 0.15259136259555817,
|
| 29 |
+
"eval_runtime": 320.8429,
|
| 30 |
+
"eval_samples_per_second": 6.234,
|
| 31 |
+
"eval_steps_per_second": 1.558,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.096,
|
| 36 |
+
"grad_norm": 409.5372314453125,
|
| 37 |
+
"learning_rate": 9.511111111111111e-07,
|
| 38 |
+
"loss": 0.8091524505615234,
|
| 39 |
+
"step": 150
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.128,
|
| 43 |
+
"grad_norm": 4.935479640960693,
|
| 44 |
+
"learning_rate": 8.955555555555555e-07,
|
| 45 |
+
"loss": 0.9019821166992188,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.128,
|
| 50 |
+
"eval_loss": 0.1397811472415924,
|
| 51 |
+
"eval_runtime": 320.5384,
|
| 52 |
+
"eval_samples_per_second": 6.24,
|
| 53 |
+
"eval_steps_per_second": 1.56,
|
| 54 |
+
"step": 200
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.16,
|
| 58 |
+
"grad_norm": 2.68009877204895,
|
| 59 |
+
"learning_rate": 8.399999999999999e-07,
|
| 60 |
+
"loss": 0.8488493347167969,
|
| 61 |
+
"step": 250
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.192,
|
| 65 |
+
"grad_norm": 250.0933380126953,
|
| 66 |
+
"learning_rate": 7.844444444444445e-07,
|
| 67 |
+
"loss": 0.8013427734375,
|
| 68 |
+
"step": 300
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.192,
|
| 72 |
+
"eval_loss": 0.13730598986148834,
|
| 73 |
+
"eval_runtime": 320.7296,
|
| 74 |
+
"eval_samples_per_second": 6.236,
|
| 75 |
+
"eval_steps_per_second": 1.559,
|
| 76 |
+
"step": 300
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.224,
|
| 80 |
+
"grad_norm": 3.164937973022461,
|
| 81 |
+
"learning_rate": 7.288888888888889e-07,
|
| 82 |
+
"loss": 0.6347025299072265,
|
| 83 |
+
"step": 350
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.256,
|
| 87 |
+
"grad_norm": 120.01831817626953,
|
| 88 |
+
"learning_rate": 6.733333333333333e-07,
|
| 89 |
+
"loss": 0.960202407836914,
|
| 90 |
+
"step": 400
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.256,
|
| 94 |
+
"eval_loss": 0.12297496199607849,
|
| 95 |
+
"eval_runtime": 320.4414,
|
| 96 |
+
"eval_samples_per_second": 6.241,
|
| 97 |
+
"eval_steps_per_second": 1.56,
|
| 98 |
+
"step": 400
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.288,
|
| 102 |
+
"grad_norm": 25.20623779296875,
|
| 103 |
+
"learning_rate": 6.177777777777777e-07,
|
| 104 |
+
"loss": 0.7631096649169922,
|
| 105 |
+
"step": 450
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.32,
|
| 109 |
+
"grad_norm": 124.12076568603516,
|
| 110 |
+
"learning_rate": 5.622222222222222e-07,
|
| 111 |
+
"loss": 0.9197311401367188,
|
| 112 |
+
"step": 500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.32,
|
| 116 |
+
"eval_loss": 0.13506844639778137,
|
| 117 |
+
"eval_runtime": 320.4909,
|
| 118 |
+
"eval_samples_per_second": 6.24,
|
| 119 |
+
"eval_steps_per_second": 1.56,
|
| 120 |
+
"step": 500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.352,
|
| 124 |
+
"grad_norm": 24.420225143432617,
|
| 125 |
+
"learning_rate": 5.066666666666667e-07,
|
| 126 |
+
"loss": 0.7606878662109375,
|
| 127 |
+
"step": 550
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.384,
|
| 131 |
+
"grad_norm": 354.97039794921875,
|
| 132 |
+
"learning_rate": 4.511111111111111e-07,
|
| 133 |
+
"loss": 0.7917160034179688,
|
| 134 |
+
"step": 600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.384,
|
| 138 |
+
"eval_loss": 0.14342211186885834,
|
| 139 |
+
"eval_runtime": 313.4668,
|
| 140 |
+
"eval_samples_per_second": 6.38,
|
| 141 |
+
"eval_steps_per_second": 1.595,
|
| 142 |
+
"step": 600
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.416,
|
| 146 |
+
"grad_norm": 10.307374000549316,
|
| 147 |
+
"learning_rate": 3.955555555555555e-07,
|
| 148 |
+
"loss": 0.7462944030761719,
|
| 149 |
+
"step": 650
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.448,
|
| 153 |
+
"grad_norm": 318.738525390625,
|
| 154 |
+
"learning_rate": 3.4000000000000003e-07,
|
| 155 |
+
"loss": 0.7443762969970703,
|
| 156 |
+
"step": 700
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.448,
|
| 160 |
+
"eval_loss": 0.11785667389631271,
|
| 161 |
+
"eval_runtime": 320.2436,
|
| 162 |
+
"eval_samples_per_second": 6.245,
|
| 163 |
+
"eval_steps_per_second": 1.561,
|
| 164 |
+
"step": 700
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.48,
|
| 168 |
+
"grad_norm": 21.945383071899414,
|
| 169 |
+
"learning_rate": 2.8555555555555556e-07,
|
| 170 |
+
"loss": 0.8074851989746094,
|
| 171 |
+
"step": 750
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.512,
|
| 175 |
+
"grad_norm": 22.927444458007812,
|
| 176 |
+
"learning_rate": 2.3e-07,
|
| 177 |
+
"loss": 0.8185167694091797,
|
| 178 |
+
"step": 800
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.512,
|
| 182 |
+
"eval_loss": 0.12330672889947891,
|
| 183 |
+
"eval_runtime": 320.7694,
|
| 184 |
+
"eval_samples_per_second": 6.235,
|
| 185 |
+
"eval_steps_per_second": 1.559,
|
| 186 |
+
"step": 800
|
| 187 |
+
}
|
| 188 |
+
],
|
| 189 |
+
"logging_steps": 50,
|
| 190 |
+
"max_steps": 1000,
|
| 191 |
+
"num_input_tokens_seen": 0,
|
| 192 |
+
"num_train_epochs": 1,
|
| 193 |
+
"save_steps": 100,
|
| 194 |
+
"stateful_callbacks": {
|
| 195 |
+
"TrainerControl": {
|
| 196 |
+
"args": {
|
| 197 |
+
"should_epoch_stop": false,
|
| 198 |
+
"should_evaluate": false,
|
| 199 |
+
"should_log": false,
|
| 200 |
+
"should_save": true,
|
| 201 |
+
"should_training_stop": false
|
| 202 |
+
},
|
| 203 |
+
"attributes": {}
|
| 204 |
+
}
|
| 205 |
+
},
|
| 206 |
+
"total_flos": 4.760771474107181e+16,
|
| 207 |
+
"train_batch_size": 4,
|
| 208 |
+
"trial_name": null,
|
| 209 |
+
"trial_params": null
|
| 210 |
+
}
|
checkpoint-800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3da6e2a61b6bad49c09f23bbfa4f7ef9e3d7901027f4ad3d2b43105b61dd0d2d
|
| 3 |
+
size 4792
|
checkpoint-900/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"DebertaV2ForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": null,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"eos_token_id": null,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 4096,
|
| 14 |
+
"layer_norm_eps": 1e-07,
|
| 15 |
+
"legacy": true,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"max_relative_positions": -1,
|
| 18 |
+
"model_type": "deberta-v2",
|
| 19 |
+
"norm_rel_ebd": "layer_norm",
|
| 20 |
+
"num_attention_heads": 16,
|
| 21 |
+
"num_hidden_layers": 24,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"pooler_dropout": 0,
|
| 24 |
+
"pooler_hidden_act": "gelu",
|
| 25 |
+
"pooler_hidden_size": 1024,
|
| 26 |
+
"pos_att_type": [
|
| 27 |
+
"p2c",
|
| 28 |
+
"c2p"
|
| 29 |
+
],
|
| 30 |
+
"position_biased_input": false,
|
| 31 |
+
"position_buckets": 256,
|
| 32 |
+
"relative_attention": true,
|
| 33 |
+
"share_att_key": true,
|
| 34 |
+
"tie_word_embeddings": true,
|
| 35 |
+
"transformers_version": "5.12.1",
|
| 36 |
+
"type_vocab_size": 0,
|
| 37 |
+
"use_cache": false,
|
| 38 |
+
"vocab_size": 128100
|
| 39 |
+
}
|
checkpoint-900/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2865eb6acb03ed325c073c2adaf7271c952711d542ac947b0c51e99796024edd
|
| 3 |
+
size 1740304440
|
checkpoint-900/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d676fecd8cc8c8f2a24537538ec3d7de0b129bdd002a5ead7516f2960d59b6c
|
| 3 |
+
size 2271441914
|
checkpoint-900/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93efbee84c08fb59d1f3798c4b23d8c8b6a806fd75c4b0964b7f14bc28e1ae7e
|
| 3 |
+
size 14180
|
checkpoint-900/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8431376bf95d1c0546c1eef3cb47a3f11ee296eab2c473fa31691df1225983e
|
| 3 |
+
size 988
|
checkpoint-900/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:453a00e2bfc50195cd3457c699c8cd37686801ab0ed4c4e49b0a8692dc8c44ea
|
| 3 |
+
size 1064
|
checkpoint-900/trainer_state.json
ADDED
|
@@ -0,0 +1,232 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.576,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 900,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 25.09992027282715,
|
| 15 |
+
"learning_rate": 4.3999999999999997e-07,
|
| 16 |
+
"loss": 1.0546456146240235,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.064,
|
| 21 |
+
"grad_norm": 44.00643539428711,
|
| 22 |
+
"learning_rate": 9.399999999999999e-07,
|
| 23 |
+
"loss": 1.2045201873779297,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.064,
|
| 28 |
+
"eval_loss": 0.15259136259555817,
|
| 29 |
+
"eval_runtime": 320.8429,
|
| 30 |
+
"eval_samples_per_second": 6.234,
|
| 31 |
+
"eval_steps_per_second": 1.558,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.096,
|
| 36 |
+
"grad_norm": 409.5372314453125,
|
| 37 |
+
"learning_rate": 9.511111111111111e-07,
|
| 38 |
+
"loss": 0.8091524505615234,
|
| 39 |
+
"step": 150
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.128,
|
| 43 |
+
"grad_norm": 4.935479640960693,
|
| 44 |
+
"learning_rate": 8.955555555555555e-07,
|
| 45 |
+
"loss": 0.9019821166992188,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.128,
|
| 50 |
+
"eval_loss": 0.1397811472415924,
|
| 51 |
+
"eval_runtime": 320.5384,
|
| 52 |
+
"eval_samples_per_second": 6.24,
|
| 53 |
+
"eval_steps_per_second": 1.56,
|
| 54 |
+
"step": 200
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.16,
|
| 58 |
+
"grad_norm": 2.68009877204895,
|
| 59 |
+
"learning_rate": 8.399999999999999e-07,
|
| 60 |
+
"loss": 0.8488493347167969,
|
| 61 |
+
"step": 250
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.192,
|
| 65 |
+
"grad_norm": 250.0933380126953,
|
| 66 |
+
"learning_rate": 7.844444444444445e-07,
|
| 67 |
+
"loss": 0.8013427734375,
|
| 68 |
+
"step": 300
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.192,
|
| 72 |
+
"eval_loss": 0.13730598986148834,
|
| 73 |
+
"eval_runtime": 320.7296,
|
| 74 |
+
"eval_samples_per_second": 6.236,
|
| 75 |
+
"eval_steps_per_second": 1.559,
|
| 76 |
+
"step": 300
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.224,
|
| 80 |
+
"grad_norm": 3.164937973022461,
|
| 81 |
+
"learning_rate": 7.288888888888889e-07,
|
| 82 |
+
"loss": 0.6347025299072265,
|
| 83 |
+
"step": 350
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.256,
|
| 87 |
+
"grad_norm": 120.01831817626953,
|
| 88 |
+
"learning_rate": 6.733333333333333e-07,
|
| 89 |
+
"loss": 0.960202407836914,
|
| 90 |
+
"step": 400
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.256,
|
| 94 |
+
"eval_loss": 0.12297496199607849,
|
| 95 |
+
"eval_runtime": 320.4414,
|
| 96 |
+
"eval_samples_per_second": 6.241,
|
| 97 |
+
"eval_steps_per_second": 1.56,
|
| 98 |
+
"step": 400
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.288,
|
| 102 |
+
"grad_norm": 25.20623779296875,
|
| 103 |
+
"learning_rate": 6.177777777777777e-07,
|
| 104 |
+
"loss": 0.7631096649169922,
|
| 105 |
+
"step": 450
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.32,
|
| 109 |
+
"grad_norm": 124.12076568603516,
|
| 110 |
+
"learning_rate": 5.622222222222222e-07,
|
| 111 |
+
"loss": 0.9197311401367188,
|
| 112 |
+
"step": 500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.32,
|
| 116 |
+
"eval_loss": 0.13506844639778137,
|
| 117 |
+
"eval_runtime": 320.4909,
|
| 118 |
+
"eval_samples_per_second": 6.24,
|
| 119 |
+
"eval_steps_per_second": 1.56,
|
| 120 |
+
"step": 500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.352,
|
| 124 |
+
"grad_norm": 24.420225143432617,
|
| 125 |
+
"learning_rate": 5.066666666666667e-07,
|
| 126 |
+
"loss": 0.7606878662109375,
|
| 127 |
+
"step": 550
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.384,
|
| 131 |
+
"grad_norm": 354.97039794921875,
|
| 132 |
+
"learning_rate": 4.511111111111111e-07,
|
| 133 |
+
"loss": 0.7917160034179688,
|
| 134 |
+
"step": 600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.384,
|
| 138 |
+
"eval_loss": 0.14342211186885834,
|
| 139 |
+
"eval_runtime": 313.4668,
|
| 140 |
+
"eval_samples_per_second": 6.38,
|
| 141 |
+
"eval_steps_per_second": 1.595,
|
| 142 |
+
"step": 600
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.416,
|
| 146 |
+
"grad_norm": 10.307374000549316,
|
| 147 |
+
"learning_rate": 3.955555555555555e-07,
|
| 148 |
+
"loss": 0.7462944030761719,
|
| 149 |
+
"step": 650
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.448,
|
| 153 |
+
"grad_norm": 318.738525390625,
|
| 154 |
+
"learning_rate": 3.4000000000000003e-07,
|
| 155 |
+
"loss": 0.7443762969970703,
|
| 156 |
+
"step": 700
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.448,
|
| 160 |
+
"eval_loss": 0.11785667389631271,
|
| 161 |
+
"eval_runtime": 320.2436,
|
| 162 |
+
"eval_samples_per_second": 6.245,
|
| 163 |
+
"eval_steps_per_second": 1.561,
|
| 164 |
+
"step": 700
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.48,
|
| 168 |
+
"grad_norm": 21.945383071899414,
|
| 169 |
+
"learning_rate": 2.8555555555555556e-07,
|
| 170 |
+
"loss": 0.8074851989746094,
|
| 171 |
+
"step": 750
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.512,
|
| 175 |
+
"grad_norm": 22.927444458007812,
|
| 176 |
+
"learning_rate": 2.3e-07,
|
| 177 |
+
"loss": 0.8185167694091797,
|
| 178 |
+
"step": 800
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.512,
|
| 182 |
+
"eval_loss": 0.12330672889947891,
|
| 183 |
+
"eval_runtime": 320.7694,
|
| 184 |
+
"eval_samples_per_second": 6.235,
|
| 185 |
+
"eval_steps_per_second": 1.559,
|
| 186 |
+
"step": 800
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.544,
|
| 190 |
+
"grad_norm": 124.74988555908203,
|
| 191 |
+
"learning_rate": 1.7444444444444445e-07,
|
| 192 |
+
"loss": 0.9554645538330078,
|
| 193 |
+
"step": 850
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.576,
|
| 197 |
+
"grad_norm": 445.3316650390625,
|
| 198 |
+
"learning_rate": 1.1888888888888889e-07,
|
| 199 |
+
"loss": 1.1002864837646484,
|
| 200 |
+
"step": 900
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.576,
|
| 204 |
+
"eval_loss": 0.1164385974407196,
|
| 205 |
+
"eval_runtime": 320.5865,
|
| 206 |
+
"eval_samples_per_second": 6.239,
|
| 207 |
+
"eval_steps_per_second": 1.56,
|
| 208 |
+
"step": 900
|
| 209 |
+
}
|
| 210 |
+
],
|
| 211 |
+
"logging_steps": 50,
|
| 212 |
+
"max_steps": 1000,
|
| 213 |
+
"num_input_tokens_seen": 0,
|
| 214 |
+
"num_train_epochs": 1,
|
| 215 |
+
"save_steps": 100,
|
| 216 |
+
"stateful_callbacks": {
|
| 217 |
+
"TrainerControl": {
|
| 218 |
+
"args": {
|
| 219 |
+
"should_epoch_stop": false,
|
| 220 |
+
"should_evaluate": false,
|
| 221 |
+
"should_log": false,
|
| 222 |
+
"should_save": true,
|
| 223 |
+
"should_training_stop": false
|
| 224 |
+
},
|
| 225 |
+
"attributes": {}
|
| 226 |
+
}
|
| 227 |
+
},
|
| 228 |
+
"total_flos": 5.35611144989052e+16,
|
| 229 |
+
"train_batch_size": 4,
|
| 230 |
+
"trial_name": null,
|
| 231 |
+
"trial_params": null
|
| 232 |
+
}
|
checkpoint-900/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3da6e2a61b6bad49c09f23bbfa4f7ef9e3d7901027f4ad3d2b43105b61dd0d2d
|
| 3 |
+
size 4792
|