Upload folder using huggingface_hub
Browse files- checkpoint-1140/config.json +42 -0
- checkpoint-1140/model.safetensors +3 -0
- checkpoint-1140/optimizer.pt +3 -0
- checkpoint-1140/rng_state.pth +3 -0
- checkpoint-1140/scheduler.pt +3 -0
- checkpoint-1140/trainer_state.json +210 -0
- checkpoint-1140/training_args.bin +3 -0
- checkpoint-1710/config.json +42 -0
- checkpoint-1710/model.safetensors +3 -0
- checkpoint-1710/optimizer.pt +3 -0
- checkpoint-1710/rng_state.pth +3 -0
- checkpoint-1710/scheduler.pt +3 -0
- checkpoint-1710/trainer_state.json +305 -0
- checkpoint-1710/training_args.bin +3 -0
- checkpoint-2280/config.json +42 -0
- checkpoint-2280/model.safetensors +3 -0
- checkpoint-2280/optimizer.pt +3 -0
- checkpoint-2280/rng_state.pth +3 -0
- checkpoint-2280/scheduler.pt +3 -0
- checkpoint-2280/trainer_state.json +393 -0
- checkpoint-2280/training_args.bin +3 -0
- checkpoint-2850/config.json +42 -0
- checkpoint-2850/model.safetensors +3 -0
- checkpoint-2850/optimizer.pt +3 -0
- checkpoint-2850/rng_state.pth +3 -0
- checkpoint-2850/scheduler.pt +3 -0
- checkpoint-2850/trainer_state.json +488 -0
- checkpoint-2850/training_args.bin +3 -0
- checkpoint-570/config.json +42 -0
- checkpoint-570/model.safetensors +3 -0
- checkpoint-570/optimizer.pt +3 -0
- checkpoint-570/rng_state.pth +3 -0
- checkpoint-570/scheduler.pt +3 -0
- checkpoint-570/trainer_state.json +122 -0
- checkpoint-570/training_args.bin +3 -0
checkpoint-1140/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Personal",
|
| 13 |
+
"1": "Political",
|
| 14 |
+
"2": "Religious",
|
| 15 |
+
"3": "Geopolitical"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"0": "Personal",
|
| 21 |
+
"1": "Political",
|
| 22 |
+
"2": "Religious",
|
| 23 |
+
"3": "Geopolitical"
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "electra",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"summary_activation": "gelu",
|
| 34 |
+
"summary_last_dropout": 0.1,
|
| 35 |
+
"summary_type": "first",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.52.4",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
checkpoint-1140/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c9beb9d87d4f5718f68ba66439c0c038939098e05898cf03d296b35fea432ae
|
| 3 |
+
size 442505824
|
checkpoint-1140/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a4b08f8ca8f22a289ae4c8ebc2e9aff452273afe0011b30ea2d8d755cddccb6
|
| 3 |
+
size 885131514
|
checkpoint-1140/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff5b9afe2c551807c3590bab42bdaef11b8025fda7e72d48d393d1fcc4144903
|
| 3 |
+
size 14244
|
checkpoint-1140/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:efe11eee4a6bb9a0bce4feef531b6f4b82c136a07c5b5c3ceaa90c5ea36b78ed
|
| 3 |
+
size 1064
|
checkpoint-1140/trainer_state.json
ADDED
|
@@ -0,0 +1,210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1140,
|
| 3 |
+
"best_metric": 0.8596423207276622,
|
| 4 |
+
"best_model_checkpoint": "./banglabert-hate-speech/checkpoint-1140",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1140,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 2.613939046859741,
|
| 15 |
+
"learning_rate": 2.9484210526315792e-05,
|
| 16 |
+
"loss": 1.2847,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.17543859649122806,
|
| 21 |
+
"grad_norm": 2.230876922607422,
|
| 22 |
+
"learning_rate": 2.8957894736842105e-05,
|
| 23 |
+
"loss": 1.1609,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2631578947368421,
|
| 28 |
+
"grad_norm": 10.591499328613281,
|
| 29 |
+
"learning_rate": 2.8431578947368422e-05,
|
| 30 |
+
"loss": 1.033,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3508771929824561,
|
| 35 |
+
"grad_norm": 4.717245101928711,
|
| 36 |
+
"learning_rate": 2.7905263157894738e-05,
|
| 37 |
+
"loss": 0.9866,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43859649122807015,
|
| 42 |
+
"grad_norm": 7.096415042877197,
|
| 43 |
+
"learning_rate": 2.7378947368421055e-05,
|
| 44 |
+
"loss": 0.9235,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5263157894736842,
|
| 49 |
+
"grad_norm": 5.946366786956787,
|
| 50 |
+
"learning_rate": 2.6852631578947368e-05,
|
| 51 |
+
"loss": 0.7276,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.6140350877192983,
|
| 56 |
+
"grad_norm": 2.292181968688965,
|
| 57 |
+
"learning_rate": 2.6326315789473687e-05,
|
| 58 |
+
"loss": 0.6928,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.7017543859649122,
|
| 63 |
+
"grad_norm": 7.117995262145996,
|
| 64 |
+
"learning_rate": 2.58e-05,
|
| 65 |
+
"loss": 0.7734,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7894736842105263,
|
| 70 |
+
"grad_norm": 1.2756074666976929,
|
| 71 |
+
"learning_rate": 2.5273684210526317e-05,
|
| 72 |
+
"loss": 0.602,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.8771929824561403,
|
| 77 |
+
"grad_norm": 2.6861534118652344,
|
| 78 |
+
"learning_rate": 2.4747368421052633e-05,
|
| 79 |
+
"loss": 0.6134,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.9649122807017544,
|
| 84 |
+
"grad_norm": 11.638345718383789,
|
| 85 |
+
"learning_rate": 2.422105263157895e-05,
|
| 86 |
+
"loss": 0.517,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.8491228070175438,
|
| 92 |
+
"eval_f1_macro": 0.8364544727451704,
|
| 93 |
+
"eval_f1_weighted": 0.8507649468201081,
|
| 94 |
+
"eval_loss": 0.501762330532074,
|
| 95 |
+
"eval_runtime": 9.2272,
|
| 96 |
+
"eval_samples_per_second": 61.774,
|
| 97 |
+
"eval_steps_per_second": 7.803,
|
| 98 |
+
"step": 570
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0526315789473684,
|
| 102 |
+
"grad_norm": 7.222433090209961,
|
| 103 |
+
"learning_rate": 2.3694736842105262e-05,
|
| 104 |
+
"loss": 0.4792,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.1403508771929824,
|
| 109 |
+
"grad_norm": 14.4037504196167,
|
| 110 |
+
"learning_rate": 2.3168421052631582e-05,
|
| 111 |
+
"loss": 0.4212,
|
| 112 |
+
"step": 650
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.2280701754385965,
|
| 116 |
+
"grad_norm": 1.6871181726455688,
|
| 117 |
+
"learning_rate": 2.2642105263157895e-05,
|
| 118 |
+
"loss": 0.5112,
|
| 119 |
+
"step": 700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.3157894736842106,
|
| 123 |
+
"grad_norm": 12.315402030944824,
|
| 124 |
+
"learning_rate": 2.211578947368421e-05,
|
| 125 |
+
"loss": 0.3467,
|
| 126 |
+
"step": 750
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.4035087719298245,
|
| 130 |
+
"grad_norm": 26.74939727783203,
|
| 131 |
+
"learning_rate": 2.1589473684210528e-05,
|
| 132 |
+
"loss": 0.5507,
|
| 133 |
+
"step": 800
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.4912280701754386,
|
| 137 |
+
"grad_norm": 11.844022750854492,
|
| 138 |
+
"learning_rate": 2.1063157894736844e-05,
|
| 139 |
+
"loss": 0.4319,
|
| 140 |
+
"step": 850
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.5789473684210527,
|
| 144 |
+
"grad_norm": 0.30326738953590393,
|
| 145 |
+
"learning_rate": 2.0536842105263157e-05,
|
| 146 |
+
"loss": 0.4583,
|
| 147 |
+
"step": 900
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.6666666666666665,
|
| 151 |
+
"grad_norm": 6.0051140785217285,
|
| 152 |
+
"learning_rate": 2.0010526315789477e-05,
|
| 153 |
+
"loss": 0.3868,
|
| 154 |
+
"step": 950
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.7543859649122808,
|
| 158 |
+
"grad_norm": 1.0601614713668823,
|
| 159 |
+
"learning_rate": 1.948421052631579e-05,
|
| 160 |
+
"loss": 0.4607,
|
| 161 |
+
"step": 1000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.8421052631578947,
|
| 165 |
+
"grad_norm": 0.5424970984458923,
|
| 166 |
+
"learning_rate": 1.8957894736842106e-05,
|
| 167 |
+
"loss": 0.4468,
|
| 168 |
+
"step": 1050
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.9298245614035088,
|
| 172 |
+
"grad_norm": 14.800076484680176,
|
| 173 |
+
"learning_rate": 1.8431578947368423e-05,
|
| 174 |
+
"loss": 0.4958,
|
| 175 |
+
"step": 1100
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.0,
|
| 179 |
+
"eval_accuracy": 0.8719298245614036,
|
| 180 |
+
"eval_f1_macro": 0.8596423207276622,
|
| 181 |
+
"eval_f1_weighted": 0.8710256158201644,
|
| 182 |
+
"eval_loss": 0.48058807849884033,
|
| 183 |
+
"eval_runtime": 9.1284,
|
| 184 |
+
"eval_samples_per_second": 62.442,
|
| 185 |
+
"eval_steps_per_second": 7.887,
|
| 186 |
+
"step": 1140
|
| 187 |
+
}
|
| 188 |
+
],
|
| 189 |
+
"logging_steps": 50,
|
| 190 |
+
"max_steps": 2850,
|
| 191 |
+
"num_input_tokens_seen": 0,
|
| 192 |
+
"num_train_epochs": 5,
|
| 193 |
+
"save_steps": 500,
|
| 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": 1199281725677568.0,
|
| 207 |
+
"train_batch_size": 8,
|
| 208 |
+
"trial_name": null,
|
| 209 |
+
"trial_params": null
|
| 210 |
+
}
|
checkpoint-1140/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75e59a4485b50670550ee1f98c4f17155f387568fb8c6b921b878e2ea24de3f8
|
| 3 |
+
size 5304
|
checkpoint-1710/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Personal",
|
| 13 |
+
"1": "Political",
|
| 14 |
+
"2": "Religious",
|
| 15 |
+
"3": "Geopolitical"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"0": "Personal",
|
| 21 |
+
"1": "Political",
|
| 22 |
+
"2": "Religious",
|
| 23 |
+
"3": "Geopolitical"
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "electra",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"summary_activation": "gelu",
|
| 34 |
+
"summary_last_dropout": 0.1,
|
| 35 |
+
"summary_type": "first",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.52.4",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
checkpoint-1710/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2913bbcde2c37bf69585c41308a91aa354cba8447ae2f82909b05eb92bb87897
|
| 3 |
+
size 442505824
|
checkpoint-1710/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4eee0cdec76ad0d1b1a3927635d35cef35151c6625da91e250c8e1c9b6e3635
|
| 3 |
+
size 885131514
|
checkpoint-1710/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ef2ed34cefa3c64aafe034252aabaf5bc4a526e5fb0d06ef6d466a1df11bb42
|
| 3 |
+
size 14244
|
checkpoint-1710/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d195afd31b7974adc795724099e3f38c5d8dcd84f04dc1ddd6cdf6e5ee28aee
|
| 3 |
+
size 1064
|
checkpoint-1710/trainer_state.json
ADDED
|
@@ -0,0 +1,305 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1710,
|
| 3 |
+
"best_metric": 0.876431567492838,
|
| 4 |
+
"best_model_checkpoint": "./banglabert-hate-speech/checkpoint-1710",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1710,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 2.613939046859741,
|
| 15 |
+
"learning_rate": 2.9484210526315792e-05,
|
| 16 |
+
"loss": 1.2847,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.17543859649122806,
|
| 21 |
+
"grad_norm": 2.230876922607422,
|
| 22 |
+
"learning_rate": 2.8957894736842105e-05,
|
| 23 |
+
"loss": 1.1609,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2631578947368421,
|
| 28 |
+
"grad_norm": 10.591499328613281,
|
| 29 |
+
"learning_rate": 2.8431578947368422e-05,
|
| 30 |
+
"loss": 1.033,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3508771929824561,
|
| 35 |
+
"grad_norm": 4.717245101928711,
|
| 36 |
+
"learning_rate": 2.7905263157894738e-05,
|
| 37 |
+
"loss": 0.9866,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43859649122807015,
|
| 42 |
+
"grad_norm": 7.096415042877197,
|
| 43 |
+
"learning_rate": 2.7378947368421055e-05,
|
| 44 |
+
"loss": 0.9235,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5263157894736842,
|
| 49 |
+
"grad_norm": 5.946366786956787,
|
| 50 |
+
"learning_rate": 2.6852631578947368e-05,
|
| 51 |
+
"loss": 0.7276,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.6140350877192983,
|
| 56 |
+
"grad_norm": 2.292181968688965,
|
| 57 |
+
"learning_rate": 2.6326315789473687e-05,
|
| 58 |
+
"loss": 0.6928,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.7017543859649122,
|
| 63 |
+
"grad_norm": 7.117995262145996,
|
| 64 |
+
"learning_rate": 2.58e-05,
|
| 65 |
+
"loss": 0.7734,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7894736842105263,
|
| 70 |
+
"grad_norm": 1.2756074666976929,
|
| 71 |
+
"learning_rate": 2.5273684210526317e-05,
|
| 72 |
+
"loss": 0.602,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.8771929824561403,
|
| 77 |
+
"grad_norm": 2.6861534118652344,
|
| 78 |
+
"learning_rate": 2.4747368421052633e-05,
|
| 79 |
+
"loss": 0.6134,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.9649122807017544,
|
| 84 |
+
"grad_norm": 11.638345718383789,
|
| 85 |
+
"learning_rate": 2.422105263157895e-05,
|
| 86 |
+
"loss": 0.517,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.8491228070175438,
|
| 92 |
+
"eval_f1_macro": 0.8364544727451704,
|
| 93 |
+
"eval_f1_weighted": 0.8507649468201081,
|
| 94 |
+
"eval_loss": 0.501762330532074,
|
| 95 |
+
"eval_runtime": 9.2272,
|
| 96 |
+
"eval_samples_per_second": 61.774,
|
| 97 |
+
"eval_steps_per_second": 7.803,
|
| 98 |
+
"step": 570
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0526315789473684,
|
| 102 |
+
"grad_norm": 7.222433090209961,
|
| 103 |
+
"learning_rate": 2.3694736842105262e-05,
|
| 104 |
+
"loss": 0.4792,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.1403508771929824,
|
| 109 |
+
"grad_norm": 14.4037504196167,
|
| 110 |
+
"learning_rate": 2.3168421052631582e-05,
|
| 111 |
+
"loss": 0.4212,
|
| 112 |
+
"step": 650
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.2280701754385965,
|
| 116 |
+
"grad_norm": 1.6871181726455688,
|
| 117 |
+
"learning_rate": 2.2642105263157895e-05,
|
| 118 |
+
"loss": 0.5112,
|
| 119 |
+
"step": 700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.3157894736842106,
|
| 123 |
+
"grad_norm": 12.315402030944824,
|
| 124 |
+
"learning_rate": 2.211578947368421e-05,
|
| 125 |
+
"loss": 0.3467,
|
| 126 |
+
"step": 750
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.4035087719298245,
|
| 130 |
+
"grad_norm": 26.74939727783203,
|
| 131 |
+
"learning_rate": 2.1589473684210528e-05,
|
| 132 |
+
"loss": 0.5507,
|
| 133 |
+
"step": 800
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.4912280701754386,
|
| 137 |
+
"grad_norm": 11.844022750854492,
|
| 138 |
+
"learning_rate": 2.1063157894736844e-05,
|
| 139 |
+
"loss": 0.4319,
|
| 140 |
+
"step": 850
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.5789473684210527,
|
| 144 |
+
"grad_norm": 0.30326738953590393,
|
| 145 |
+
"learning_rate": 2.0536842105263157e-05,
|
| 146 |
+
"loss": 0.4583,
|
| 147 |
+
"step": 900
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.6666666666666665,
|
| 151 |
+
"grad_norm": 6.0051140785217285,
|
| 152 |
+
"learning_rate": 2.0010526315789477e-05,
|
| 153 |
+
"loss": 0.3868,
|
| 154 |
+
"step": 950
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.7543859649122808,
|
| 158 |
+
"grad_norm": 1.0601614713668823,
|
| 159 |
+
"learning_rate": 1.948421052631579e-05,
|
| 160 |
+
"loss": 0.4607,
|
| 161 |
+
"step": 1000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.8421052631578947,
|
| 165 |
+
"grad_norm": 0.5424970984458923,
|
| 166 |
+
"learning_rate": 1.8957894736842106e-05,
|
| 167 |
+
"loss": 0.4468,
|
| 168 |
+
"step": 1050
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.9298245614035088,
|
| 172 |
+
"grad_norm": 14.800076484680176,
|
| 173 |
+
"learning_rate": 1.8431578947368423e-05,
|
| 174 |
+
"loss": 0.4958,
|
| 175 |
+
"step": 1100
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.0,
|
| 179 |
+
"eval_accuracy": 0.8719298245614036,
|
| 180 |
+
"eval_f1_macro": 0.8596423207276622,
|
| 181 |
+
"eval_f1_weighted": 0.8710256158201644,
|
| 182 |
+
"eval_loss": 0.48058807849884033,
|
| 183 |
+
"eval_runtime": 9.1284,
|
| 184 |
+
"eval_samples_per_second": 62.442,
|
| 185 |
+
"eval_steps_per_second": 7.887,
|
| 186 |
+
"step": 1140
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.017543859649123,
|
| 190 |
+
"grad_norm": 20.266742706298828,
|
| 191 |
+
"learning_rate": 1.7905263157894736e-05,
|
| 192 |
+
"loss": 0.3943,
|
| 193 |
+
"step": 1150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.1052631578947367,
|
| 197 |
+
"grad_norm": 0.22429589927196503,
|
| 198 |
+
"learning_rate": 1.7378947368421052e-05,
|
| 199 |
+
"loss": 0.2996,
|
| 200 |
+
"step": 1200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.192982456140351,
|
| 204 |
+
"grad_norm": 0.8915501236915588,
|
| 205 |
+
"learning_rate": 1.685263157894737e-05,
|
| 206 |
+
"loss": 0.3662,
|
| 207 |
+
"step": 1250
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.280701754385965,
|
| 211 |
+
"grad_norm": 1.173509955406189,
|
| 212 |
+
"learning_rate": 1.6326315789473685e-05,
|
| 213 |
+
"loss": 0.2995,
|
| 214 |
+
"step": 1300
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.3684210526315788,
|
| 218 |
+
"grad_norm": 0.16873787343502045,
|
| 219 |
+
"learning_rate": 1.5799999999999998e-05,
|
| 220 |
+
"loss": 0.3123,
|
| 221 |
+
"step": 1350
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.456140350877193,
|
| 225 |
+
"grad_norm": 13.355467796325684,
|
| 226 |
+
"learning_rate": 1.5273684210526318e-05,
|
| 227 |
+
"loss": 0.3127,
|
| 228 |
+
"step": 1400
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.543859649122807,
|
| 232 |
+
"grad_norm": 13.341830253601074,
|
| 233 |
+
"learning_rate": 1.4747368421052632e-05,
|
| 234 |
+
"loss": 0.2525,
|
| 235 |
+
"step": 1450
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.6315789473684212,
|
| 239 |
+
"grad_norm": 0.2885662913322449,
|
| 240 |
+
"learning_rate": 1.4221052631578949e-05,
|
| 241 |
+
"loss": 0.2998,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.719298245614035,
|
| 246 |
+
"grad_norm": 2.877472400665283,
|
| 247 |
+
"learning_rate": 1.3694736842105263e-05,
|
| 248 |
+
"loss": 0.3174,
|
| 249 |
+
"step": 1550
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.807017543859649,
|
| 253 |
+
"grad_norm": 3.695666790008545,
|
| 254 |
+
"learning_rate": 1.316842105263158e-05,
|
| 255 |
+
"loss": 0.341,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.8947368421052633,
|
| 260 |
+
"grad_norm": 20.929218292236328,
|
| 261 |
+
"learning_rate": 1.2642105263157896e-05,
|
| 262 |
+
"loss": 0.3093,
|
| 263 |
+
"step": 1650
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.982456140350877,
|
| 267 |
+
"grad_norm": 0.13824953138828278,
|
| 268 |
+
"learning_rate": 1.211578947368421e-05,
|
| 269 |
+
"loss": 0.2969,
|
| 270 |
+
"step": 1700
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.0,
|
| 274 |
+
"eval_accuracy": 0.8859649122807017,
|
| 275 |
+
"eval_f1_macro": 0.876431567492838,
|
| 276 |
+
"eval_f1_weighted": 0.8855810649898733,
|
| 277 |
+
"eval_loss": 0.5191295742988586,
|
| 278 |
+
"eval_runtime": 9.2195,
|
| 279 |
+
"eval_samples_per_second": 61.826,
|
| 280 |
+
"eval_steps_per_second": 7.81,
|
| 281 |
+
"step": 1710
|
| 282 |
+
}
|
| 283 |
+
],
|
| 284 |
+
"logging_steps": 50,
|
| 285 |
+
"max_steps": 2850,
|
| 286 |
+
"num_input_tokens_seen": 0,
|
| 287 |
+
"num_train_epochs": 5,
|
| 288 |
+
"save_steps": 500,
|
| 289 |
+
"stateful_callbacks": {
|
| 290 |
+
"TrainerControl": {
|
| 291 |
+
"args": {
|
| 292 |
+
"should_epoch_stop": false,
|
| 293 |
+
"should_evaluate": false,
|
| 294 |
+
"should_log": false,
|
| 295 |
+
"should_save": true,
|
| 296 |
+
"should_training_stop": false
|
| 297 |
+
},
|
| 298 |
+
"attributes": {}
|
| 299 |
+
}
|
| 300 |
+
},
|
| 301 |
+
"total_flos": 1798922588516352.0,
|
| 302 |
+
"train_batch_size": 8,
|
| 303 |
+
"trial_name": null,
|
| 304 |
+
"trial_params": null
|
| 305 |
+
}
|
checkpoint-1710/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75e59a4485b50670550ee1f98c4f17155f387568fb8c6b921b878e2ea24de3f8
|
| 3 |
+
size 5304
|
checkpoint-2280/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Personal",
|
| 13 |
+
"1": "Political",
|
| 14 |
+
"2": "Religious",
|
| 15 |
+
"3": "Geopolitical"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"0": "Personal",
|
| 21 |
+
"1": "Political",
|
| 22 |
+
"2": "Religious",
|
| 23 |
+
"3": "Geopolitical"
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "electra",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"summary_activation": "gelu",
|
| 34 |
+
"summary_last_dropout": 0.1,
|
| 35 |
+
"summary_type": "first",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.52.4",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
checkpoint-2280/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f53cf907ff8e5ecc13cb997f3873e9ff3925f31edb36ffd1fc40375e314ae61e
|
| 3 |
+
size 442505824
|
checkpoint-2280/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db2b21ecc5fa604e0ae728d50149e1841e8dad6e932b4f61664be88bbb14a789
|
| 3 |
+
size 885131514
|
checkpoint-2280/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb163261fd9ce3891ff2abe6b12d4585aab20f6b0e4e0ad8d8d83aeb48ac480b
|
| 3 |
+
size 14244
|
checkpoint-2280/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe981a7c7d2a16d26df56c8d97387cc5dad6bcddcbf70d958184abb073cbdfaf
|
| 3 |
+
size 1064
|
checkpoint-2280/trainer_state.json
ADDED
|
@@ -0,0 +1,393 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1710,
|
| 3 |
+
"best_metric": 0.876431567492838,
|
| 4 |
+
"best_model_checkpoint": "./banglabert-hate-speech/checkpoint-1710",
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2280,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 2.613939046859741,
|
| 15 |
+
"learning_rate": 2.9484210526315792e-05,
|
| 16 |
+
"loss": 1.2847,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.17543859649122806,
|
| 21 |
+
"grad_norm": 2.230876922607422,
|
| 22 |
+
"learning_rate": 2.8957894736842105e-05,
|
| 23 |
+
"loss": 1.1609,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2631578947368421,
|
| 28 |
+
"grad_norm": 10.591499328613281,
|
| 29 |
+
"learning_rate": 2.8431578947368422e-05,
|
| 30 |
+
"loss": 1.033,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3508771929824561,
|
| 35 |
+
"grad_norm": 4.717245101928711,
|
| 36 |
+
"learning_rate": 2.7905263157894738e-05,
|
| 37 |
+
"loss": 0.9866,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43859649122807015,
|
| 42 |
+
"grad_norm": 7.096415042877197,
|
| 43 |
+
"learning_rate": 2.7378947368421055e-05,
|
| 44 |
+
"loss": 0.9235,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5263157894736842,
|
| 49 |
+
"grad_norm": 5.946366786956787,
|
| 50 |
+
"learning_rate": 2.6852631578947368e-05,
|
| 51 |
+
"loss": 0.7276,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.6140350877192983,
|
| 56 |
+
"grad_norm": 2.292181968688965,
|
| 57 |
+
"learning_rate": 2.6326315789473687e-05,
|
| 58 |
+
"loss": 0.6928,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.7017543859649122,
|
| 63 |
+
"grad_norm": 7.117995262145996,
|
| 64 |
+
"learning_rate": 2.58e-05,
|
| 65 |
+
"loss": 0.7734,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7894736842105263,
|
| 70 |
+
"grad_norm": 1.2756074666976929,
|
| 71 |
+
"learning_rate": 2.5273684210526317e-05,
|
| 72 |
+
"loss": 0.602,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.8771929824561403,
|
| 77 |
+
"grad_norm": 2.6861534118652344,
|
| 78 |
+
"learning_rate": 2.4747368421052633e-05,
|
| 79 |
+
"loss": 0.6134,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.9649122807017544,
|
| 84 |
+
"grad_norm": 11.638345718383789,
|
| 85 |
+
"learning_rate": 2.422105263157895e-05,
|
| 86 |
+
"loss": 0.517,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.8491228070175438,
|
| 92 |
+
"eval_f1_macro": 0.8364544727451704,
|
| 93 |
+
"eval_f1_weighted": 0.8507649468201081,
|
| 94 |
+
"eval_loss": 0.501762330532074,
|
| 95 |
+
"eval_runtime": 9.2272,
|
| 96 |
+
"eval_samples_per_second": 61.774,
|
| 97 |
+
"eval_steps_per_second": 7.803,
|
| 98 |
+
"step": 570
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0526315789473684,
|
| 102 |
+
"grad_norm": 7.222433090209961,
|
| 103 |
+
"learning_rate": 2.3694736842105262e-05,
|
| 104 |
+
"loss": 0.4792,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.1403508771929824,
|
| 109 |
+
"grad_norm": 14.4037504196167,
|
| 110 |
+
"learning_rate": 2.3168421052631582e-05,
|
| 111 |
+
"loss": 0.4212,
|
| 112 |
+
"step": 650
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.2280701754385965,
|
| 116 |
+
"grad_norm": 1.6871181726455688,
|
| 117 |
+
"learning_rate": 2.2642105263157895e-05,
|
| 118 |
+
"loss": 0.5112,
|
| 119 |
+
"step": 700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.3157894736842106,
|
| 123 |
+
"grad_norm": 12.315402030944824,
|
| 124 |
+
"learning_rate": 2.211578947368421e-05,
|
| 125 |
+
"loss": 0.3467,
|
| 126 |
+
"step": 750
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.4035087719298245,
|
| 130 |
+
"grad_norm": 26.74939727783203,
|
| 131 |
+
"learning_rate": 2.1589473684210528e-05,
|
| 132 |
+
"loss": 0.5507,
|
| 133 |
+
"step": 800
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.4912280701754386,
|
| 137 |
+
"grad_norm": 11.844022750854492,
|
| 138 |
+
"learning_rate": 2.1063157894736844e-05,
|
| 139 |
+
"loss": 0.4319,
|
| 140 |
+
"step": 850
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.5789473684210527,
|
| 144 |
+
"grad_norm": 0.30326738953590393,
|
| 145 |
+
"learning_rate": 2.0536842105263157e-05,
|
| 146 |
+
"loss": 0.4583,
|
| 147 |
+
"step": 900
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.6666666666666665,
|
| 151 |
+
"grad_norm": 6.0051140785217285,
|
| 152 |
+
"learning_rate": 2.0010526315789477e-05,
|
| 153 |
+
"loss": 0.3868,
|
| 154 |
+
"step": 950
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.7543859649122808,
|
| 158 |
+
"grad_norm": 1.0601614713668823,
|
| 159 |
+
"learning_rate": 1.948421052631579e-05,
|
| 160 |
+
"loss": 0.4607,
|
| 161 |
+
"step": 1000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.8421052631578947,
|
| 165 |
+
"grad_norm": 0.5424970984458923,
|
| 166 |
+
"learning_rate": 1.8957894736842106e-05,
|
| 167 |
+
"loss": 0.4468,
|
| 168 |
+
"step": 1050
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.9298245614035088,
|
| 172 |
+
"grad_norm": 14.800076484680176,
|
| 173 |
+
"learning_rate": 1.8431578947368423e-05,
|
| 174 |
+
"loss": 0.4958,
|
| 175 |
+
"step": 1100
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.0,
|
| 179 |
+
"eval_accuracy": 0.8719298245614036,
|
| 180 |
+
"eval_f1_macro": 0.8596423207276622,
|
| 181 |
+
"eval_f1_weighted": 0.8710256158201644,
|
| 182 |
+
"eval_loss": 0.48058807849884033,
|
| 183 |
+
"eval_runtime": 9.1284,
|
| 184 |
+
"eval_samples_per_second": 62.442,
|
| 185 |
+
"eval_steps_per_second": 7.887,
|
| 186 |
+
"step": 1140
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.017543859649123,
|
| 190 |
+
"grad_norm": 20.266742706298828,
|
| 191 |
+
"learning_rate": 1.7905263157894736e-05,
|
| 192 |
+
"loss": 0.3943,
|
| 193 |
+
"step": 1150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.1052631578947367,
|
| 197 |
+
"grad_norm": 0.22429589927196503,
|
| 198 |
+
"learning_rate": 1.7378947368421052e-05,
|
| 199 |
+
"loss": 0.2996,
|
| 200 |
+
"step": 1200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.192982456140351,
|
| 204 |
+
"grad_norm": 0.8915501236915588,
|
| 205 |
+
"learning_rate": 1.685263157894737e-05,
|
| 206 |
+
"loss": 0.3662,
|
| 207 |
+
"step": 1250
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.280701754385965,
|
| 211 |
+
"grad_norm": 1.173509955406189,
|
| 212 |
+
"learning_rate": 1.6326315789473685e-05,
|
| 213 |
+
"loss": 0.2995,
|
| 214 |
+
"step": 1300
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.3684210526315788,
|
| 218 |
+
"grad_norm": 0.16873787343502045,
|
| 219 |
+
"learning_rate": 1.5799999999999998e-05,
|
| 220 |
+
"loss": 0.3123,
|
| 221 |
+
"step": 1350
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.456140350877193,
|
| 225 |
+
"grad_norm": 13.355467796325684,
|
| 226 |
+
"learning_rate": 1.5273684210526318e-05,
|
| 227 |
+
"loss": 0.3127,
|
| 228 |
+
"step": 1400
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.543859649122807,
|
| 232 |
+
"grad_norm": 13.341830253601074,
|
| 233 |
+
"learning_rate": 1.4747368421052632e-05,
|
| 234 |
+
"loss": 0.2525,
|
| 235 |
+
"step": 1450
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.6315789473684212,
|
| 239 |
+
"grad_norm": 0.2885662913322449,
|
| 240 |
+
"learning_rate": 1.4221052631578949e-05,
|
| 241 |
+
"loss": 0.2998,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.719298245614035,
|
| 246 |
+
"grad_norm": 2.877472400665283,
|
| 247 |
+
"learning_rate": 1.3694736842105263e-05,
|
| 248 |
+
"loss": 0.3174,
|
| 249 |
+
"step": 1550
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.807017543859649,
|
| 253 |
+
"grad_norm": 3.695666790008545,
|
| 254 |
+
"learning_rate": 1.316842105263158e-05,
|
| 255 |
+
"loss": 0.341,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.8947368421052633,
|
| 260 |
+
"grad_norm": 20.929218292236328,
|
| 261 |
+
"learning_rate": 1.2642105263157896e-05,
|
| 262 |
+
"loss": 0.3093,
|
| 263 |
+
"step": 1650
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.982456140350877,
|
| 267 |
+
"grad_norm": 0.13824953138828278,
|
| 268 |
+
"learning_rate": 1.211578947368421e-05,
|
| 269 |
+
"loss": 0.2969,
|
| 270 |
+
"step": 1700
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.0,
|
| 274 |
+
"eval_accuracy": 0.8859649122807017,
|
| 275 |
+
"eval_f1_macro": 0.876431567492838,
|
| 276 |
+
"eval_f1_weighted": 0.8855810649898733,
|
| 277 |
+
"eval_loss": 0.5191295742988586,
|
| 278 |
+
"eval_runtime": 9.2195,
|
| 279 |
+
"eval_samples_per_second": 61.826,
|
| 280 |
+
"eval_steps_per_second": 7.81,
|
| 281 |
+
"step": 1710
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.0701754385964914,
|
| 285 |
+
"grad_norm": 0.19890473783016205,
|
| 286 |
+
"learning_rate": 1.1589473684210527e-05,
|
| 287 |
+
"loss": 0.2249,
|
| 288 |
+
"step": 1750
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.1578947368421053,
|
| 292 |
+
"grad_norm": 0.09896814078092575,
|
| 293 |
+
"learning_rate": 1.1063157894736843e-05,
|
| 294 |
+
"loss": 0.1578,
|
| 295 |
+
"step": 1800
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.245614035087719,
|
| 299 |
+
"grad_norm": 0.18045368790626526,
|
| 300 |
+
"learning_rate": 1.0536842105263158e-05,
|
| 301 |
+
"loss": 0.1534,
|
| 302 |
+
"step": 1850
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.3333333333333335,
|
| 306 |
+
"grad_norm": 2.493330478668213,
|
| 307 |
+
"learning_rate": 1.0010526315789474e-05,
|
| 308 |
+
"loss": 0.181,
|
| 309 |
+
"step": 1900
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.4210526315789473,
|
| 313 |
+
"grad_norm": 0.25888559222221375,
|
| 314 |
+
"learning_rate": 9.484210526315791e-06,
|
| 315 |
+
"loss": 0.1758,
|
| 316 |
+
"step": 1950
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.5087719298245617,
|
| 320 |
+
"grad_norm": 59.44745635986328,
|
| 321 |
+
"learning_rate": 8.957894736842106e-06,
|
| 322 |
+
"loss": 0.2045,
|
| 323 |
+
"step": 2000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.5964912280701755,
|
| 327 |
+
"grad_norm": 0.15124382078647614,
|
| 328 |
+
"learning_rate": 8.431578947368422e-06,
|
| 329 |
+
"loss": 0.3103,
|
| 330 |
+
"step": 2050
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.6842105263157894,
|
| 334 |
+
"grad_norm": 4.733994007110596,
|
| 335 |
+
"learning_rate": 7.905263157894738e-06,
|
| 336 |
+
"loss": 0.2384,
|
| 337 |
+
"step": 2100
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.7719298245614032,
|
| 341 |
+
"grad_norm": 0.46031907200813293,
|
| 342 |
+
"learning_rate": 7.378947368421053e-06,
|
| 343 |
+
"loss": 0.1849,
|
| 344 |
+
"step": 2150
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.8596491228070176,
|
| 348 |
+
"grad_norm": 5.210213661193848,
|
| 349 |
+
"learning_rate": 6.8526315789473685e-06,
|
| 350 |
+
"loss": 0.2268,
|
| 351 |
+
"step": 2200
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.9473684210526314,
|
| 355 |
+
"grad_norm": 0.19484597444534302,
|
| 356 |
+
"learning_rate": 6.326315789473684e-06,
|
| 357 |
+
"loss": 0.2313,
|
| 358 |
+
"step": 2250
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 4.0,
|
| 362 |
+
"eval_accuracy": 0.875438596491228,
|
| 363 |
+
"eval_f1_macro": 0.8628016579647876,
|
| 364 |
+
"eval_f1_weighted": 0.8751100908173923,
|
| 365 |
+
"eval_loss": 0.5712200999259949,
|
| 366 |
+
"eval_runtime": 9.187,
|
| 367 |
+
"eval_samples_per_second": 62.044,
|
| 368 |
+
"eval_steps_per_second": 7.837,
|
| 369 |
+
"step": 2280
|
| 370 |
+
}
|
| 371 |
+
],
|
| 372 |
+
"logging_steps": 50,
|
| 373 |
+
"max_steps": 2850,
|
| 374 |
+
"num_input_tokens_seen": 0,
|
| 375 |
+
"num_train_epochs": 5,
|
| 376 |
+
"save_steps": 500,
|
| 377 |
+
"stateful_callbacks": {
|
| 378 |
+
"TrainerControl": {
|
| 379 |
+
"args": {
|
| 380 |
+
"should_epoch_stop": false,
|
| 381 |
+
"should_evaluate": false,
|
| 382 |
+
"should_log": false,
|
| 383 |
+
"should_save": true,
|
| 384 |
+
"should_training_stop": false
|
| 385 |
+
},
|
| 386 |
+
"attributes": {}
|
| 387 |
+
}
|
| 388 |
+
},
|
| 389 |
+
"total_flos": 2398563451355136.0,
|
| 390 |
+
"train_batch_size": 8,
|
| 391 |
+
"trial_name": null,
|
| 392 |
+
"trial_params": null
|
| 393 |
+
}
|
checkpoint-2280/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75e59a4485b50670550ee1f98c4f17155f387568fb8c6b921b878e2ea24de3f8
|
| 3 |
+
size 5304
|
checkpoint-2850/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Personal",
|
| 13 |
+
"1": "Political",
|
| 14 |
+
"2": "Religious",
|
| 15 |
+
"3": "Geopolitical"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"0": "Personal",
|
| 21 |
+
"1": "Political",
|
| 22 |
+
"2": "Religious",
|
| 23 |
+
"3": "Geopolitical"
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "electra",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"summary_activation": "gelu",
|
| 34 |
+
"summary_last_dropout": 0.1,
|
| 35 |
+
"summary_type": "first",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.52.4",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
checkpoint-2850/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:103e82756deba030dabe6c6131e9c35ddd75617542c822bcb88d9d47ea3d692c
|
| 3 |
+
size 442505824
|
checkpoint-2850/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fba9d38022474d3dbcb755e4c864d65ebba0ad98dad2735698de8a1d53a15834
|
| 3 |
+
size 885131514
|
checkpoint-2850/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42a34d67eebeddc815e3e11e61db97d051e4d58700b1bfcc8001f211636083fb
|
| 3 |
+
size 14244
|
checkpoint-2850/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:469267c99727d972deab65a94ded1577aa272e0ff46092f59ffd1d373c2c738d
|
| 3 |
+
size 1064
|
checkpoint-2850/trainer_state.json
ADDED
|
@@ -0,0 +1,488 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1710,
|
| 3 |
+
"best_metric": 0.876431567492838,
|
| 4 |
+
"best_model_checkpoint": "./banglabert-hate-speech/checkpoint-1710",
|
| 5 |
+
"epoch": 5.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2850,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 2.613939046859741,
|
| 15 |
+
"learning_rate": 2.9484210526315792e-05,
|
| 16 |
+
"loss": 1.2847,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.17543859649122806,
|
| 21 |
+
"grad_norm": 2.230876922607422,
|
| 22 |
+
"learning_rate": 2.8957894736842105e-05,
|
| 23 |
+
"loss": 1.1609,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2631578947368421,
|
| 28 |
+
"grad_norm": 10.591499328613281,
|
| 29 |
+
"learning_rate": 2.8431578947368422e-05,
|
| 30 |
+
"loss": 1.033,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3508771929824561,
|
| 35 |
+
"grad_norm": 4.717245101928711,
|
| 36 |
+
"learning_rate": 2.7905263157894738e-05,
|
| 37 |
+
"loss": 0.9866,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43859649122807015,
|
| 42 |
+
"grad_norm": 7.096415042877197,
|
| 43 |
+
"learning_rate": 2.7378947368421055e-05,
|
| 44 |
+
"loss": 0.9235,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5263157894736842,
|
| 49 |
+
"grad_norm": 5.946366786956787,
|
| 50 |
+
"learning_rate": 2.6852631578947368e-05,
|
| 51 |
+
"loss": 0.7276,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.6140350877192983,
|
| 56 |
+
"grad_norm": 2.292181968688965,
|
| 57 |
+
"learning_rate": 2.6326315789473687e-05,
|
| 58 |
+
"loss": 0.6928,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.7017543859649122,
|
| 63 |
+
"grad_norm": 7.117995262145996,
|
| 64 |
+
"learning_rate": 2.58e-05,
|
| 65 |
+
"loss": 0.7734,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7894736842105263,
|
| 70 |
+
"grad_norm": 1.2756074666976929,
|
| 71 |
+
"learning_rate": 2.5273684210526317e-05,
|
| 72 |
+
"loss": 0.602,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.8771929824561403,
|
| 77 |
+
"grad_norm": 2.6861534118652344,
|
| 78 |
+
"learning_rate": 2.4747368421052633e-05,
|
| 79 |
+
"loss": 0.6134,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.9649122807017544,
|
| 84 |
+
"grad_norm": 11.638345718383789,
|
| 85 |
+
"learning_rate": 2.422105263157895e-05,
|
| 86 |
+
"loss": 0.517,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.8491228070175438,
|
| 92 |
+
"eval_f1_macro": 0.8364544727451704,
|
| 93 |
+
"eval_f1_weighted": 0.8507649468201081,
|
| 94 |
+
"eval_loss": 0.501762330532074,
|
| 95 |
+
"eval_runtime": 9.2272,
|
| 96 |
+
"eval_samples_per_second": 61.774,
|
| 97 |
+
"eval_steps_per_second": 7.803,
|
| 98 |
+
"step": 570
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0526315789473684,
|
| 102 |
+
"grad_norm": 7.222433090209961,
|
| 103 |
+
"learning_rate": 2.3694736842105262e-05,
|
| 104 |
+
"loss": 0.4792,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.1403508771929824,
|
| 109 |
+
"grad_norm": 14.4037504196167,
|
| 110 |
+
"learning_rate": 2.3168421052631582e-05,
|
| 111 |
+
"loss": 0.4212,
|
| 112 |
+
"step": 650
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.2280701754385965,
|
| 116 |
+
"grad_norm": 1.6871181726455688,
|
| 117 |
+
"learning_rate": 2.2642105263157895e-05,
|
| 118 |
+
"loss": 0.5112,
|
| 119 |
+
"step": 700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.3157894736842106,
|
| 123 |
+
"grad_norm": 12.315402030944824,
|
| 124 |
+
"learning_rate": 2.211578947368421e-05,
|
| 125 |
+
"loss": 0.3467,
|
| 126 |
+
"step": 750
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.4035087719298245,
|
| 130 |
+
"grad_norm": 26.74939727783203,
|
| 131 |
+
"learning_rate": 2.1589473684210528e-05,
|
| 132 |
+
"loss": 0.5507,
|
| 133 |
+
"step": 800
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.4912280701754386,
|
| 137 |
+
"grad_norm": 11.844022750854492,
|
| 138 |
+
"learning_rate": 2.1063157894736844e-05,
|
| 139 |
+
"loss": 0.4319,
|
| 140 |
+
"step": 850
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.5789473684210527,
|
| 144 |
+
"grad_norm": 0.30326738953590393,
|
| 145 |
+
"learning_rate": 2.0536842105263157e-05,
|
| 146 |
+
"loss": 0.4583,
|
| 147 |
+
"step": 900
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.6666666666666665,
|
| 151 |
+
"grad_norm": 6.0051140785217285,
|
| 152 |
+
"learning_rate": 2.0010526315789477e-05,
|
| 153 |
+
"loss": 0.3868,
|
| 154 |
+
"step": 950
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.7543859649122808,
|
| 158 |
+
"grad_norm": 1.0601614713668823,
|
| 159 |
+
"learning_rate": 1.948421052631579e-05,
|
| 160 |
+
"loss": 0.4607,
|
| 161 |
+
"step": 1000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.8421052631578947,
|
| 165 |
+
"grad_norm": 0.5424970984458923,
|
| 166 |
+
"learning_rate": 1.8957894736842106e-05,
|
| 167 |
+
"loss": 0.4468,
|
| 168 |
+
"step": 1050
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.9298245614035088,
|
| 172 |
+
"grad_norm": 14.800076484680176,
|
| 173 |
+
"learning_rate": 1.8431578947368423e-05,
|
| 174 |
+
"loss": 0.4958,
|
| 175 |
+
"step": 1100
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.0,
|
| 179 |
+
"eval_accuracy": 0.8719298245614036,
|
| 180 |
+
"eval_f1_macro": 0.8596423207276622,
|
| 181 |
+
"eval_f1_weighted": 0.8710256158201644,
|
| 182 |
+
"eval_loss": 0.48058807849884033,
|
| 183 |
+
"eval_runtime": 9.1284,
|
| 184 |
+
"eval_samples_per_second": 62.442,
|
| 185 |
+
"eval_steps_per_second": 7.887,
|
| 186 |
+
"step": 1140
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.017543859649123,
|
| 190 |
+
"grad_norm": 20.266742706298828,
|
| 191 |
+
"learning_rate": 1.7905263157894736e-05,
|
| 192 |
+
"loss": 0.3943,
|
| 193 |
+
"step": 1150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.1052631578947367,
|
| 197 |
+
"grad_norm": 0.22429589927196503,
|
| 198 |
+
"learning_rate": 1.7378947368421052e-05,
|
| 199 |
+
"loss": 0.2996,
|
| 200 |
+
"step": 1200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.192982456140351,
|
| 204 |
+
"grad_norm": 0.8915501236915588,
|
| 205 |
+
"learning_rate": 1.685263157894737e-05,
|
| 206 |
+
"loss": 0.3662,
|
| 207 |
+
"step": 1250
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.280701754385965,
|
| 211 |
+
"grad_norm": 1.173509955406189,
|
| 212 |
+
"learning_rate": 1.6326315789473685e-05,
|
| 213 |
+
"loss": 0.2995,
|
| 214 |
+
"step": 1300
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.3684210526315788,
|
| 218 |
+
"grad_norm": 0.16873787343502045,
|
| 219 |
+
"learning_rate": 1.5799999999999998e-05,
|
| 220 |
+
"loss": 0.3123,
|
| 221 |
+
"step": 1350
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.456140350877193,
|
| 225 |
+
"grad_norm": 13.355467796325684,
|
| 226 |
+
"learning_rate": 1.5273684210526318e-05,
|
| 227 |
+
"loss": 0.3127,
|
| 228 |
+
"step": 1400
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.543859649122807,
|
| 232 |
+
"grad_norm": 13.341830253601074,
|
| 233 |
+
"learning_rate": 1.4747368421052632e-05,
|
| 234 |
+
"loss": 0.2525,
|
| 235 |
+
"step": 1450
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.6315789473684212,
|
| 239 |
+
"grad_norm": 0.2885662913322449,
|
| 240 |
+
"learning_rate": 1.4221052631578949e-05,
|
| 241 |
+
"loss": 0.2998,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.719298245614035,
|
| 246 |
+
"grad_norm": 2.877472400665283,
|
| 247 |
+
"learning_rate": 1.3694736842105263e-05,
|
| 248 |
+
"loss": 0.3174,
|
| 249 |
+
"step": 1550
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.807017543859649,
|
| 253 |
+
"grad_norm": 3.695666790008545,
|
| 254 |
+
"learning_rate": 1.316842105263158e-05,
|
| 255 |
+
"loss": 0.341,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.8947368421052633,
|
| 260 |
+
"grad_norm": 20.929218292236328,
|
| 261 |
+
"learning_rate": 1.2642105263157896e-05,
|
| 262 |
+
"loss": 0.3093,
|
| 263 |
+
"step": 1650
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.982456140350877,
|
| 267 |
+
"grad_norm": 0.13824953138828278,
|
| 268 |
+
"learning_rate": 1.211578947368421e-05,
|
| 269 |
+
"loss": 0.2969,
|
| 270 |
+
"step": 1700
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.0,
|
| 274 |
+
"eval_accuracy": 0.8859649122807017,
|
| 275 |
+
"eval_f1_macro": 0.876431567492838,
|
| 276 |
+
"eval_f1_weighted": 0.8855810649898733,
|
| 277 |
+
"eval_loss": 0.5191295742988586,
|
| 278 |
+
"eval_runtime": 9.2195,
|
| 279 |
+
"eval_samples_per_second": 61.826,
|
| 280 |
+
"eval_steps_per_second": 7.81,
|
| 281 |
+
"step": 1710
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.0701754385964914,
|
| 285 |
+
"grad_norm": 0.19890473783016205,
|
| 286 |
+
"learning_rate": 1.1589473684210527e-05,
|
| 287 |
+
"loss": 0.2249,
|
| 288 |
+
"step": 1750
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.1578947368421053,
|
| 292 |
+
"grad_norm": 0.09896814078092575,
|
| 293 |
+
"learning_rate": 1.1063157894736843e-05,
|
| 294 |
+
"loss": 0.1578,
|
| 295 |
+
"step": 1800
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.245614035087719,
|
| 299 |
+
"grad_norm": 0.18045368790626526,
|
| 300 |
+
"learning_rate": 1.0536842105263158e-05,
|
| 301 |
+
"loss": 0.1534,
|
| 302 |
+
"step": 1850
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.3333333333333335,
|
| 306 |
+
"grad_norm": 2.493330478668213,
|
| 307 |
+
"learning_rate": 1.0010526315789474e-05,
|
| 308 |
+
"loss": 0.181,
|
| 309 |
+
"step": 1900
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.4210526315789473,
|
| 313 |
+
"grad_norm": 0.25888559222221375,
|
| 314 |
+
"learning_rate": 9.484210526315791e-06,
|
| 315 |
+
"loss": 0.1758,
|
| 316 |
+
"step": 1950
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.5087719298245617,
|
| 320 |
+
"grad_norm": 59.44745635986328,
|
| 321 |
+
"learning_rate": 8.957894736842106e-06,
|
| 322 |
+
"loss": 0.2045,
|
| 323 |
+
"step": 2000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.5964912280701755,
|
| 327 |
+
"grad_norm": 0.15124382078647614,
|
| 328 |
+
"learning_rate": 8.431578947368422e-06,
|
| 329 |
+
"loss": 0.3103,
|
| 330 |
+
"step": 2050
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.6842105263157894,
|
| 334 |
+
"grad_norm": 4.733994007110596,
|
| 335 |
+
"learning_rate": 7.905263157894738e-06,
|
| 336 |
+
"loss": 0.2384,
|
| 337 |
+
"step": 2100
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.7719298245614032,
|
| 341 |
+
"grad_norm": 0.46031907200813293,
|
| 342 |
+
"learning_rate": 7.378947368421053e-06,
|
| 343 |
+
"loss": 0.1849,
|
| 344 |
+
"step": 2150
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.8596491228070176,
|
| 348 |
+
"grad_norm": 5.210213661193848,
|
| 349 |
+
"learning_rate": 6.8526315789473685e-06,
|
| 350 |
+
"loss": 0.2268,
|
| 351 |
+
"step": 2200
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.9473684210526314,
|
| 355 |
+
"grad_norm": 0.19484597444534302,
|
| 356 |
+
"learning_rate": 6.326315789473684e-06,
|
| 357 |
+
"loss": 0.2313,
|
| 358 |
+
"step": 2250
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 4.0,
|
| 362 |
+
"eval_accuracy": 0.875438596491228,
|
| 363 |
+
"eval_f1_macro": 0.8628016579647876,
|
| 364 |
+
"eval_f1_weighted": 0.8751100908173923,
|
| 365 |
+
"eval_loss": 0.5712200999259949,
|
| 366 |
+
"eval_runtime": 9.187,
|
| 367 |
+
"eval_samples_per_second": 62.044,
|
| 368 |
+
"eval_steps_per_second": 7.837,
|
| 369 |
+
"step": 2280
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 4.035087719298246,
|
| 373 |
+
"grad_norm": 0.6245447397232056,
|
| 374 |
+
"learning_rate": 5.8e-06,
|
| 375 |
+
"loss": 0.1366,
|
| 376 |
+
"step": 2300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 4.12280701754386,
|
| 380 |
+
"grad_norm": 13.598281860351562,
|
| 381 |
+
"learning_rate": 5.273684210526316e-06,
|
| 382 |
+
"loss": 0.1524,
|
| 383 |
+
"step": 2350
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.2105263157894735,
|
| 387 |
+
"grad_norm": 0.09752348810434341,
|
| 388 |
+
"learning_rate": 4.747368421052631e-06,
|
| 389 |
+
"loss": 0.1104,
|
| 390 |
+
"step": 2400
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 4.298245614035087,
|
| 394 |
+
"grad_norm": 0.12292918562889099,
|
| 395 |
+
"learning_rate": 4.221052631578948e-06,
|
| 396 |
+
"loss": 0.1157,
|
| 397 |
+
"step": 2450
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 4.385964912280702,
|
| 401 |
+
"grad_norm": 0.06472612172365189,
|
| 402 |
+
"learning_rate": 3.6947368421052633e-06,
|
| 403 |
+
"loss": 0.1327,
|
| 404 |
+
"step": 2500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.473684210526316,
|
| 408 |
+
"grad_norm": 3.8604917526245117,
|
| 409 |
+
"learning_rate": 3.1684210526315793e-06,
|
| 410 |
+
"loss": 0.1394,
|
| 411 |
+
"step": 2550
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.56140350877193,
|
| 415 |
+
"grad_norm": 0.390594482421875,
|
| 416 |
+
"learning_rate": 2.6421052631578948e-06,
|
| 417 |
+
"loss": 0.1442,
|
| 418 |
+
"step": 2600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 4.649122807017544,
|
| 422 |
+
"grad_norm": 1.7911845445632935,
|
| 423 |
+
"learning_rate": 2.1157894736842107e-06,
|
| 424 |
+
"loss": 0.1759,
|
| 425 |
+
"step": 2650
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 4.7368421052631575,
|
| 429 |
+
"grad_norm": 0.17604109644889832,
|
| 430 |
+
"learning_rate": 1.5894736842105263e-06,
|
| 431 |
+
"loss": 0.1394,
|
| 432 |
+
"step": 2700
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 4.824561403508772,
|
| 436 |
+
"grad_norm": 0.08516625314950943,
|
| 437 |
+
"learning_rate": 1.063157894736842e-06,
|
| 438 |
+
"loss": 0.1408,
|
| 439 |
+
"step": 2750
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 4.912280701754386,
|
| 443 |
+
"grad_norm": 0.03146979585289955,
|
| 444 |
+
"learning_rate": 5.368421052631578e-07,
|
| 445 |
+
"loss": 0.1251,
|
| 446 |
+
"step": 2800
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 5.0,
|
| 450 |
+
"grad_norm": 3.661275863647461,
|
| 451 |
+
"learning_rate": 1.0526315789473684e-08,
|
| 452 |
+
"loss": 0.0834,
|
| 453 |
+
"step": 2850
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 5.0,
|
| 457 |
+
"eval_accuracy": 0.8859649122807017,
|
| 458 |
+
"eval_f1_macro": 0.8727182063209552,
|
| 459 |
+
"eval_f1_weighted": 0.8854333132110179,
|
| 460 |
+
"eval_loss": 0.5799562931060791,
|
| 461 |
+
"eval_runtime": 9.1822,
|
| 462 |
+
"eval_samples_per_second": 62.077,
|
| 463 |
+
"eval_steps_per_second": 7.841,
|
| 464 |
+
"step": 2850
|
| 465 |
+
}
|
| 466 |
+
],
|
| 467 |
+
"logging_steps": 50,
|
| 468 |
+
"max_steps": 2850,
|
| 469 |
+
"num_input_tokens_seen": 0,
|
| 470 |
+
"num_train_epochs": 5,
|
| 471 |
+
"save_steps": 500,
|
| 472 |
+
"stateful_callbacks": {
|
| 473 |
+
"TrainerControl": {
|
| 474 |
+
"args": {
|
| 475 |
+
"should_epoch_stop": false,
|
| 476 |
+
"should_evaluate": false,
|
| 477 |
+
"should_log": false,
|
| 478 |
+
"should_save": true,
|
| 479 |
+
"should_training_stop": true
|
| 480 |
+
},
|
| 481 |
+
"attributes": {}
|
| 482 |
+
}
|
| 483 |
+
},
|
| 484 |
+
"total_flos": 2998204314193920.0,
|
| 485 |
+
"train_batch_size": 8,
|
| 486 |
+
"trial_name": null,
|
| 487 |
+
"trial_params": null
|
| 488 |
+
}
|
checkpoint-2850/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75e59a4485b50670550ee1f98c4f17155f387568fb8c6b921b878e2ea24de3f8
|
| 3 |
+
size 5304
|
checkpoint-570/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ElectraForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Personal",
|
| 13 |
+
"1": "Political",
|
| 14 |
+
"2": "Religious",
|
| 15 |
+
"3": "Geopolitical"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"0": "Personal",
|
| 21 |
+
"1": "Political",
|
| 22 |
+
"2": "Religious",
|
| 23 |
+
"3": "Geopolitical"
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "electra",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"summary_activation": "gelu",
|
| 34 |
+
"summary_last_dropout": 0.1,
|
| 35 |
+
"summary_type": "first",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.52.4",
|
| 39 |
+
"type_vocab_size": 2,
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 32000
|
| 42 |
+
}
|
checkpoint-570/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ac333d69f89fdc32bde19b14d4116b849c6111a5314f31613e821639d59be83
|
| 3 |
+
size 442505824
|
checkpoint-570/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:444c3b1195463a3224bb41abd49d002723211f122800308c6c074d11b2bb2f7b
|
| 3 |
+
size 885131514
|
checkpoint-570/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:497ebba5762cf7b3a63a1bb7a679a7fc15de654767c2ef5d74fe834391b49a30
|
| 3 |
+
size 14244
|
checkpoint-570/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7534b6da25b19285c5c1070dba5090606637aa8907845f7a46fec774f90bfdb
|
| 3 |
+
size 1064
|
checkpoint-570/trainer_state.json
ADDED
|
@@ -0,0 +1,122 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 570,
|
| 3 |
+
"best_metric": 0.8364544727451704,
|
| 4 |
+
"best_model_checkpoint": "./banglabert-hate-speech/checkpoint-570",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 570,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 2.613939046859741,
|
| 15 |
+
"learning_rate": 2.9484210526315792e-05,
|
| 16 |
+
"loss": 1.2847,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.17543859649122806,
|
| 21 |
+
"grad_norm": 2.230876922607422,
|
| 22 |
+
"learning_rate": 2.8957894736842105e-05,
|
| 23 |
+
"loss": 1.1609,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2631578947368421,
|
| 28 |
+
"grad_norm": 10.591499328613281,
|
| 29 |
+
"learning_rate": 2.8431578947368422e-05,
|
| 30 |
+
"loss": 1.033,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3508771929824561,
|
| 35 |
+
"grad_norm": 4.717245101928711,
|
| 36 |
+
"learning_rate": 2.7905263157894738e-05,
|
| 37 |
+
"loss": 0.9866,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43859649122807015,
|
| 42 |
+
"grad_norm": 7.096415042877197,
|
| 43 |
+
"learning_rate": 2.7378947368421055e-05,
|
| 44 |
+
"loss": 0.9235,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.5263157894736842,
|
| 49 |
+
"grad_norm": 5.946366786956787,
|
| 50 |
+
"learning_rate": 2.6852631578947368e-05,
|
| 51 |
+
"loss": 0.7276,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.6140350877192983,
|
| 56 |
+
"grad_norm": 2.292181968688965,
|
| 57 |
+
"learning_rate": 2.6326315789473687e-05,
|
| 58 |
+
"loss": 0.6928,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.7017543859649122,
|
| 63 |
+
"grad_norm": 7.117995262145996,
|
| 64 |
+
"learning_rate": 2.58e-05,
|
| 65 |
+
"loss": 0.7734,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7894736842105263,
|
| 70 |
+
"grad_norm": 1.2756074666976929,
|
| 71 |
+
"learning_rate": 2.5273684210526317e-05,
|
| 72 |
+
"loss": 0.602,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.8771929824561403,
|
| 77 |
+
"grad_norm": 2.6861534118652344,
|
| 78 |
+
"learning_rate": 2.4747368421052633e-05,
|
| 79 |
+
"loss": 0.6134,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.9649122807017544,
|
| 84 |
+
"grad_norm": 11.638345718383789,
|
| 85 |
+
"learning_rate": 2.422105263157895e-05,
|
| 86 |
+
"loss": 0.517,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_accuracy": 0.8491228070175438,
|
| 92 |
+
"eval_f1_macro": 0.8364544727451704,
|
| 93 |
+
"eval_f1_weighted": 0.8507649468201081,
|
| 94 |
+
"eval_loss": 0.501762330532074,
|
| 95 |
+
"eval_runtime": 9.2272,
|
| 96 |
+
"eval_samples_per_second": 61.774,
|
| 97 |
+
"eval_steps_per_second": 7.803,
|
| 98 |
+
"step": 570
|
| 99 |
+
}
|
| 100 |
+
],
|
| 101 |
+
"logging_steps": 50,
|
| 102 |
+
"max_steps": 2850,
|
| 103 |
+
"num_input_tokens_seen": 0,
|
| 104 |
+
"num_train_epochs": 5,
|
| 105 |
+
"save_steps": 500,
|
| 106 |
+
"stateful_callbacks": {
|
| 107 |
+
"TrainerControl": {
|
| 108 |
+
"args": {
|
| 109 |
+
"should_epoch_stop": false,
|
| 110 |
+
"should_evaluate": false,
|
| 111 |
+
"should_log": false,
|
| 112 |
+
"should_save": true,
|
| 113 |
+
"should_training_stop": false
|
| 114 |
+
},
|
| 115 |
+
"attributes": {}
|
| 116 |
+
}
|
| 117 |
+
},
|
| 118 |
+
"total_flos": 599640862838784.0,
|
| 119 |
+
"train_batch_size": 8,
|
| 120 |
+
"trial_name": null,
|
| 121 |
+
"trial_params": null
|
| 122 |
+
}
|
checkpoint-570/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75e59a4485b50670550ee1f98c4f17155f387568fb8c6b921b878e2ea24de3f8
|
| 3 |
+
size 5304
|