Upload folder using huggingface_hub
Browse files- checkpoint-6424/config.json +58 -0
- checkpoint-6424/generation_config.json +8 -0
- checkpoint-6424/model.safetensors +3 -0
- checkpoint-6424/optimizer.pt +3 -0
- checkpoint-6424/rng_state.pth +3 -0
- checkpoint-6424/scaler.pt +3 -0
- checkpoint-6424/scheduler.pt +3 -0
- checkpoint-6424/special_tokens_map.json +9 -0
- checkpoint-6424/tokenizer.json +0 -0
- checkpoint-6424/tokenizer_config.json +25 -0
- checkpoint-6424/trainer_state.json +166 -0
- checkpoint-6424/training_args.bin +3 -0
- trainer_state.json +166 -0
checkpoint-6424/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"add_bias_logits": false,
|
| 5 |
+
"add_final_layer_norm": true,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"MBartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.1,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.1,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 768,
|
| 14 |
+
"decoder_attention_heads": 12,
|
| 15 |
+
"decoder_ffn_dim": 3072,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 6,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"do_blenderbot_90_layernorm": false,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"early_stopping": null,
|
| 22 |
+
"encoder_attention_heads": 12,
|
| 23 |
+
"encoder_ffn_dim": 3072,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 6,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"extra_pos_embeddings": 2,
|
| 28 |
+
"force_bos_token_to_be_generated": false,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"max_position_embeddings": 1024,
|
| 44 |
+
"model_type": "mbart",
|
| 45 |
+
"no_repeat_ngram_size": null,
|
| 46 |
+
"normalize_before": true,
|
| 47 |
+
"normalize_embedding": true,
|
| 48 |
+
"num_beams": null,
|
| 49 |
+
"num_hidden_layers": 6,
|
| 50 |
+
"pad_token_id": 1,
|
| 51 |
+
"scale_embedding": false,
|
| 52 |
+
"static_position_embeddings": false,
|
| 53 |
+
"tokenizer_class": "BarthezTokenizer",
|
| 54 |
+
"torch_dtype": "float32",
|
| 55 |
+
"transformers_version": "4.51.3",
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 86001
|
| 58 |
+
}
|
checkpoint-6424/generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"forced_eos_token_id": 2,
|
| 6 |
+
"pad_token_id": 1,
|
| 7 |
+
"transformers_version": "4.51.3"
|
| 8 |
+
}
|
checkpoint-6424/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74bcde2a9eb993ebfa7ad2120a125469366ea72d5b50e4642e9ab73eaec297c6
|
| 3 |
+
size 667849236
|
checkpoint-6424/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c571bad76232a366e69dc7727acac84a671075a174fbcfbf6473ea547f5ef45
|
| 3 |
+
size 1964811
|
checkpoint-6424/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b5c9a321d911642f4f39e094bc71f7f606a7c0729b70323a4e97b3ed73f476b
|
| 3 |
+
size 14645
|
checkpoint-6424/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:788952817d7e20534058a75dacbdd9323070be90cfa509634b49d011404572c0
|
| 3 |
+
size 1383
|
checkpoint-6424/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a317addef3c200d7a4f262c74fdc4a21ceba55a5883e574984d5309cb5ed2ad9
|
| 3 |
+
size 1465
|
checkpoint-6424/special_tokens_map.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": {
|
| 3 |
+
"content": "[PAD]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
}
|
| 9 |
+
}
|
checkpoint-6424/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-6424/tokenizer_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"86000": {
|
| 12 |
+
"content": "[PAD]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
}
|
| 19 |
+
},
|
| 20 |
+
"clean_up_tokenization_spaces": true,
|
| 21 |
+
"extra_special_tokens": {},
|
| 22 |
+
"model_max_length": 128,
|
| 23 |
+
"pad_token": "[PAD]",
|
| 24 |
+
"tokenizer_class": "PreTrainedTokenizer"
|
| 25 |
+
}
|
checkpoint-6424/trainer_state.json
ADDED
|
@@ -0,0 +1,166 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6424,
|
| 3 |
+
"best_metric": 0.708712044715695,
|
| 4 |
+
"best_model_checkpoint": "checkpoints/checkpoint-6424",
|
| 5 |
+
"epoch": 0.9999610849515508,
|
| 6 |
+
"eval_steps": 1606,
|
| 7 |
+
"global_step": 6424,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07783009689847063,
|
| 14 |
+
"grad_norm": 0.2925865948200226,
|
| 15 |
+
"learning_rate": 4.991766654390158e-05,
|
| 16 |
+
"loss": 2.2108,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.15566019379694127,
|
| 21 |
+
"grad_norm": 0.11884481459856033,
|
| 22 |
+
"learning_rate": 4.966988962239142e-05,
|
| 23 |
+
"loss": 0.3886,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2334902906954119,
|
| 28 |
+
"grad_norm": 0.09392867982387543,
|
| 29 |
+
"learning_rate": 4.925831306092169e-05,
|
| 30 |
+
"loss": 0.2271,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2499902712378877,
|
| 35 |
+
"eval_avg": 26.749637281629344,
|
| 36 |
+
"eval_cer": 0.6640905490705347,
|
| 37 |
+
"eval_der": 78.87487905513909,
|
| 38 |
+
"eval_loss": 0.13775022327899933,
|
| 39 |
+
"eval_runtime": 1439.866,
|
| 40 |
+
"eval_samples_per_second": 94.261,
|
| 41 |
+
"eval_steps_per_second": 0.369,
|
| 42 |
+
"eval_wer": 0.7099422406784007,
|
| 43 |
+
"step": 1606
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.31132038759388253,
|
| 47 |
+
"grad_norm": 0.0717502012848854,
|
| 48 |
+
"learning_rate": 4.8685669587267704e-05,
|
| 49 |
+
"loss": 0.1657,
|
| 50 |
+
"step": 2000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3891504844923532,
|
| 54 |
+
"grad_norm": 0.0638226792216301,
|
| 55 |
+
"learning_rate": 4.795576135848184e-05,
|
| 56 |
+
"loss": 0.1334,
|
| 57 |
+
"step": 2500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.4669805813908238,
|
| 61 |
+
"grad_norm": 0.05441045016050339,
|
| 62 |
+
"learning_rate": 4.707343471586959e-05,
|
| 63 |
+
"loss": 0.1135,
|
| 64 |
+
"step": 3000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.4999805424757754,
|
| 68 |
+
"eval_avg": 26.867260983703588,
|
| 69 |
+
"eval_cer": 0.6597599235984197,
|
| 70 |
+
"eval_der": 79.24906581788363,
|
| 71 |
+
"eval_loss": 0.07722582668066025,
|
| 72 |
+
"eval_runtime": 1442.4649,
|
| 73 |
+
"eval_samples_per_second": 94.091,
|
| 74 |
+
"eval_steps_per_second": 0.368,
|
| 75 |
+
"eval_wer": 0.6929572096287142,
|
| 76 |
+
"step": 3212
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.5448106782892945,
|
| 80 |
+
"grad_norm": 0.050636373460292816,
|
| 81 |
+
"learning_rate": 4.604454800693874e-05,
|
| 82 |
+
"loss": 0.1002,
|
| 83 |
+
"step": 3500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.6226407751877651,
|
| 87 |
+
"grad_norm": 0.0581156350672245,
|
| 88 |
+
"learning_rate": 4.48759326879731e-05,
|
| 89 |
+
"loss": 0.0906,
|
| 90 |
+
"step": 4000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.7004708720862357,
|
| 94 |
+
"grad_norm": 0.050500743091106415,
|
| 95 |
+
"learning_rate": 4.3575347965496405e-05,
|
| 96 |
+
"loss": 0.0828,
|
| 97 |
+
"step": 4500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.749970813713663,
|
| 101 |
+
"eval_avg": 26.888053782081823,
|
| 102 |
+
"eval_cer": 0.6586451367564623,
|
| 103 |
+
"eval_der": 79.3192047037025,
|
| 104 |
+
"eval_loss": 0.05902250111103058,
|
| 105 |
+
"eval_runtime": 1431.2383,
|
| 106 |
+
"eval_samples_per_second": 94.829,
|
| 107 |
+
"eval_steps_per_second": 0.371,
|
| 108 |
+
"eval_wer": 0.6863115057865005,
|
| 109 |
+
"step": 4818
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.7783009689847064,
|
| 113 |
+
"grad_norm": 0.04594194516539574,
|
| 114 |
+
"learning_rate": 4.9989118673869795e-05,
|
| 115 |
+
"loss": 0.0765,
|
| 116 |
+
"step": 5000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.856131065883177,
|
| 120 |
+
"grad_norm": 0.04987065866589546,
|
| 121 |
+
"learning_rate": 4.984611214237347e-05,
|
| 122 |
+
"loss": 0.0724,
|
| 123 |
+
"step": 5500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.9339611627816476,
|
| 127 |
+
"grad_norm": 0.04362853616476059,
|
| 128 |
+
"learning_rate": 4.953813591358179e-05,
|
| 129 |
+
"loss": 0.0683,
|
| 130 |
+
"step": 6000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.9999610849515508,
|
| 134 |
+
"eval_avg": 0.708712044715695,
|
| 135 |
+
"eval_cer": 0.6546741642910598,
|
| 136 |
+
"eval_der": 0.7934032551594553,
|
| 137 |
+
"eval_loss": 0.049141544848680496,
|
| 138 |
+
"eval_runtime": 305.8639,
|
| 139 |
+
"eval_samples_per_second": 443.737,
|
| 140 |
+
"eval_steps_per_second": 1.736,
|
| 141 |
+
"eval_wer": 0.6780587146965699,
|
| 142 |
+
"step": 6424
|
| 143 |
+
}
|
| 144 |
+
],
|
| 145 |
+
"logging_steps": 500,
|
| 146 |
+
"max_steps": 19272,
|
| 147 |
+
"num_input_tokens_seen": 0,
|
| 148 |
+
"num_train_epochs": 3,
|
| 149 |
+
"save_steps": 1606,
|
| 150 |
+
"stateful_callbacks": {
|
| 151 |
+
"TrainerControl": {
|
| 152 |
+
"args": {
|
| 153 |
+
"should_epoch_stop": false,
|
| 154 |
+
"should_evaluate": false,
|
| 155 |
+
"should_log": false,
|
| 156 |
+
"should_save": true,
|
| 157 |
+
"should_training_stop": false
|
| 158 |
+
},
|
| 159 |
+
"attributes": {}
|
| 160 |
+
}
|
| 161 |
+
},
|
| 162 |
+
"total_flos": 5.013847368542454e+17,
|
| 163 |
+
"train_batch_size": 256,
|
| 164 |
+
"trial_name": null,
|
| 165 |
+
"trial_params": null
|
| 166 |
+
}
|
checkpoint-6424/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cd77b85635f40803f2a499bdb589f69718a7b080ce297cef7adac845287846e
|
| 3 |
+
size 5905
|
trainer_state.json
ADDED
|
@@ -0,0 +1,166 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6424,
|
| 3 |
+
"best_metric": 0.708712044715695,
|
| 4 |
+
"best_model_checkpoint": "checkpoints/checkpoint-6424",
|
| 5 |
+
"epoch": 0.9999610849515508,
|
| 6 |
+
"eval_steps": 1606,
|
| 7 |
+
"global_step": 6424,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07783009689847063,
|
| 14 |
+
"grad_norm": 0.2925865948200226,
|
| 15 |
+
"learning_rate": 4.991766654390158e-05,
|
| 16 |
+
"loss": 2.2108,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.15566019379694127,
|
| 21 |
+
"grad_norm": 0.11884481459856033,
|
| 22 |
+
"learning_rate": 4.966988962239142e-05,
|
| 23 |
+
"loss": 0.3886,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2334902906954119,
|
| 28 |
+
"grad_norm": 0.09392867982387543,
|
| 29 |
+
"learning_rate": 4.925831306092169e-05,
|
| 30 |
+
"loss": 0.2271,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2499902712378877,
|
| 35 |
+
"eval_avg": 26.749637281629344,
|
| 36 |
+
"eval_cer": 0.6640905490705347,
|
| 37 |
+
"eval_der": 78.87487905513909,
|
| 38 |
+
"eval_loss": 0.13775022327899933,
|
| 39 |
+
"eval_runtime": 1439.866,
|
| 40 |
+
"eval_samples_per_second": 94.261,
|
| 41 |
+
"eval_steps_per_second": 0.369,
|
| 42 |
+
"eval_wer": 0.7099422406784007,
|
| 43 |
+
"step": 1606
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.31132038759388253,
|
| 47 |
+
"grad_norm": 0.0717502012848854,
|
| 48 |
+
"learning_rate": 4.8685669587267704e-05,
|
| 49 |
+
"loss": 0.1657,
|
| 50 |
+
"step": 2000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3891504844923532,
|
| 54 |
+
"grad_norm": 0.0638226792216301,
|
| 55 |
+
"learning_rate": 4.795576135848184e-05,
|
| 56 |
+
"loss": 0.1334,
|
| 57 |
+
"step": 2500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.4669805813908238,
|
| 61 |
+
"grad_norm": 0.05441045016050339,
|
| 62 |
+
"learning_rate": 4.707343471586959e-05,
|
| 63 |
+
"loss": 0.1135,
|
| 64 |
+
"step": 3000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.4999805424757754,
|
| 68 |
+
"eval_avg": 26.867260983703588,
|
| 69 |
+
"eval_cer": 0.6597599235984197,
|
| 70 |
+
"eval_der": 79.24906581788363,
|
| 71 |
+
"eval_loss": 0.07722582668066025,
|
| 72 |
+
"eval_runtime": 1442.4649,
|
| 73 |
+
"eval_samples_per_second": 94.091,
|
| 74 |
+
"eval_steps_per_second": 0.368,
|
| 75 |
+
"eval_wer": 0.6929572096287142,
|
| 76 |
+
"step": 3212
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.5448106782892945,
|
| 80 |
+
"grad_norm": 0.050636373460292816,
|
| 81 |
+
"learning_rate": 4.604454800693874e-05,
|
| 82 |
+
"loss": 0.1002,
|
| 83 |
+
"step": 3500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.6226407751877651,
|
| 87 |
+
"grad_norm": 0.0581156350672245,
|
| 88 |
+
"learning_rate": 4.48759326879731e-05,
|
| 89 |
+
"loss": 0.0906,
|
| 90 |
+
"step": 4000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.7004708720862357,
|
| 94 |
+
"grad_norm": 0.050500743091106415,
|
| 95 |
+
"learning_rate": 4.3575347965496405e-05,
|
| 96 |
+
"loss": 0.0828,
|
| 97 |
+
"step": 4500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.749970813713663,
|
| 101 |
+
"eval_avg": 26.888053782081823,
|
| 102 |
+
"eval_cer": 0.6586451367564623,
|
| 103 |
+
"eval_der": 79.3192047037025,
|
| 104 |
+
"eval_loss": 0.05902250111103058,
|
| 105 |
+
"eval_runtime": 1431.2383,
|
| 106 |
+
"eval_samples_per_second": 94.829,
|
| 107 |
+
"eval_steps_per_second": 0.371,
|
| 108 |
+
"eval_wer": 0.6863115057865005,
|
| 109 |
+
"step": 4818
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.7783009689847064,
|
| 113 |
+
"grad_norm": 0.04594194516539574,
|
| 114 |
+
"learning_rate": 4.9989118673869795e-05,
|
| 115 |
+
"loss": 0.0765,
|
| 116 |
+
"step": 5000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.856131065883177,
|
| 120 |
+
"grad_norm": 0.04987065866589546,
|
| 121 |
+
"learning_rate": 4.984611214237347e-05,
|
| 122 |
+
"loss": 0.0724,
|
| 123 |
+
"step": 5500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.9339611627816476,
|
| 127 |
+
"grad_norm": 0.04362853616476059,
|
| 128 |
+
"learning_rate": 4.953813591358179e-05,
|
| 129 |
+
"loss": 0.0683,
|
| 130 |
+
"step": 6000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.9999610849515508,
|
| 134 |
+
"eval_avg": 0.708712044715695,
|
| 135 |
+
"eval_cer": 0.6546741642910598,
|
| 136 |
+
"eval_der": 0.7934032551594553,
|
| 137 |
+
"eval_loss": 0.049141544848680496,
|
| 138 |
+
"eval_runtime": 305.8639,
|
| 139 |
+
"eval_samples_per_second": 443.737,
|
| 140 |
+
"eval_steps_per_second": 1.736,
|
| 141 |
+
"eval_wer": 0.6780587146965699,
|
| 142 |
+
"step": 6424
|
| 143 |
+
}
|
| 144 |
+
],
|
| 145 |
+
"logging_steps": 500,
|
| 146 |
+
"max_steps": 19272,
|
| 147 |
+
"num_input_tokens_seen": 0,
|
| 148 |
+
"num_train_epochs": 3,
|
| 149 |
+
"save_steps": 1606,
|
| 150 |
+
"stateful_callbacks": {
|
| 151 |
+
"TrainerControl": {
|
| 152 |
+
"args": {
|
| 153 |
+
"should_epoch_stop": false,
|
| 154 |
+
"should_evaluate": false,
|
| 155 |
+
"should_log": false,
|
| 156 |
+
"should_save": true,
|
| 157 |
+
"should_training_stop": false
|
| 158 |
+
},
|
| 159 |
+
"attributes": {}
|
| 160 |
+
}
|
| 161 |
+
},
|
| 162 |
+
"total_flos": 5.013847368542454e+17,
|
| 163 |
+
"train_batch_size": 256,
|
| 164 |
+
"trial_name": null,
|
| 165 |
+
"trial_params": null
|
| 166 |
+
}
|