Commit
•
0201ee6
1
Parent(s):
b40bb8d
Training in progress, step 4800, checkpoint
Browse files- checkpoint-4800/config.json +109 -0
- checkpoint-4800/model.safetensors +3 -0
- checkpoint-4800/optimizer.pt +3 -0
- checkpoint-4800/preprocessor_config.json +9 -0
- checkpoint-4800/rng_state.pth +3 -0
- checkpoint-4800/scheduler.pt +3 -0
- checkpoint-4800/trainer_state.json +417 -0
- checkpoint-4800/training_args.bin +3 -0
checkpoint-4800/config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "batoula187/wav2vec2-xlsr-arabic2",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 4096,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.0,
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_min_masks": 0,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_masks": 2,
|
| 69 |
+
"mask_time_prob": 0.05,
|
| 70 |
+
"model_type": "wav2vec2",
|
| 71 |
+
"num_adapter_layers": 3,
|
| 72 |
+
"num_attention_heads": 16,
|
| 73 |
+
"num_codevector_groups": 2,
|
| 74 |
+
"num_codevectors_per_group": 320,
|
| 75 |
+
"num_conv_pos_embedding_groups": 16,
|
| 76 |
+
"num_conv_pos_embeddings": 128,
|
| 77 |
+
"num_feat_extract_layers": 7,
|
| 78 |
+
"num_hidden_layers": 24,
|
| 79 |
+
"num_negatives": 100,
|
| 80 |
+
"output_hidden_size": 1024,
|
| 81 |
+
"pad_token_id": 88,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"tdnn_dilation": [
|
| 84 |
+
1,
|
| 85 |
+
2,
|
| 86 |
+
3,
|
| 87 |
+
1,
|
| 88 |
+
1
|
| 89 |
+
],
|
| 90 |
+
"tdnn_dim": [
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
512,
|
| 95 |
+
1500
|
| 96 |
+
],
|
| 97 |
+
"tdnn_kernel": [
|
| 98 |
+
5,
|
| 99 |
+
3,
|
| 100 |
+
3,
|
| 101 |
+
1,
|
| 102 |
+
1
|
| 103 |
+
],
|
| 104 |
+
"torch_dtype": "float32",
|
| 105 |
+
"transformers_version": "4.41.2",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 91,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
checkpoint-4800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74925693742fc426f90c1638a8c8b4c896ae79b4f8daf170f529f074c52b048b
|
| 3 |
+
size 1262180580
|
checkpoint-4800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9c44965f709bbb25110634485437cfa6c636165cebca71fac8de28e0eafe100
|
| 3 |
+
size 2490905910
|
checkpoint-4800/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-4800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd689aa2544d42d2d1fcdb8557f99d69e8922cbebcfeccec276d57b2ef6879b3
|
| 3 |
+
size 14308
|
checkpoint-4800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:841e28d397fe5526838938358710e70fd9b97728638b7b8b3201dbf1c2c15873
|
| 3 |
+
size 1064
|
checkpoint-4800/trainer_state.json
ADDED
|
@@ -0,0 +1,417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.18520188331604,
|
| 3 |
+
"best_model_checkpoint": "batoula187/wav2vec2-xlsr-arabic2/checkpoint-1800",
|
| 4 |
+
"epoch": 10.14799154334038,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 4800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.42283298097251587,
|
| 13 |
+
"grad_norm": 0.43856075406074524,
|
| 14 |
+
"learning_rate": 0.00011999999999999999,
|
| 15 |
+
"loss": 0.1174,
|
| 16 |
+
"step": 200
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42283298097251587,
|
| 20 |
+
"eval_loss": 1.2291227579116821,
|
| 21 |
+
"eval_runtime": 64.2106,
|
| 22 |
+
"eval_samples_per_second": 16.321,
|
| 23 |
+
"eval_steps_per_second": 2.04,
|
| 24 |
+
"eval_wer": 0.6153393487642213,
|
| 25 |
+
"step": 200
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.8456659619450317,
|
| 29 |
+
"grad_norm": 5.825935363769531,
|
| 30 |
+
"learning_rate": 0.00023999999999999998,
|
| 31 |
+
"loss": 0.098,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.8456659619450317,
|
| 36 |
+
"eval_loss": 1.232483148574829,
|
| 37 |
+
"eval_runtime": 64.9125,
|
| 38 |
+
"eval_samples_per_second": 16.145,
|
| 39 |
+
"eval_steps_per_second": 2.018,
|
| 40 |
+
"eval_wer": 0.6275009807767752,
|
| 41 |
+
"step": 400
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.2684989429175475,
|
| 45 |
+
"grad_norm": 0.9385707378387451,
|
| 46 |
+
"learning_rate": 0.00029780861943024103,
|
| 47 |
+
"loss": 0.1301,
|
| 48 |
+
"step": 600
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.2684989429175475,
|
| 52 |
+
"eval_loss": 1.1968672275543213,
|
| 53 |
+
"eval_runtime": 63.9557,
|
| 54 |
+
"eval_samples_per_second": 16.386,
|
| 55 |
+
"eval_steps_per_second": 2.048,
|
| 56 |
+
"eval_wer": 0.6127893291486858,
|
| 57 |
+
"step": 600
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.6913319238900635,
|
| 61 |
+
"grad_norm": 0.5655494332313538,
|
| 62 |
+
"learning_rate": 0.00029346968590211834,
|
| 63 |
+
"loss": 0.1514,
|
| 64 |
+
"step": 800
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 1.6913319238900635,
|
| 68 |
+
"eval_loss": 1.2292981147766113,
|
| 69 |
+
"eval_runtime": 64.6544,
|
| 70 |
+
"eval_samples_per_second": 16.209,
|
| 71 |
+
"eval_steps_per_second": 2.026,
|
| 72 |
+
"eval_wer": 0.6488819144762652,
|
| 73 |
+
"step": 800
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.1141649048625792,
|
| 77 |
+
"grad_norm": 1.0750213861465454,
|
| 78 |
+
"learning_rate": 0.0002890869247626004,
|
| 79 |
+
"loss": 0.1494,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.1141649048625792,
|
| 84 |
+
"eval_loss": 1.3062251806259155,
|
| 85 |
+
"eval_runtime": 63.5313,
|
| 86 |
+
"eval_samples_per_second": 16.496,
|
| 87 |
+
"eval_steps_per_second": 2.062,
|
| 88 |
+
"eval_wer": 0.670066692820714,
|
| 89 |
+
"step": 1000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.536997885835095,
|
| 93 |
+
"grad_norm": 0.8127875328063965,
|
| 94 |
+
"learning_rate": 0.0002847041636230825,
|
| 95 |
+
"loss": 0.1382,
|
| 96 |
+
"step": 1200
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.536997885835095,
|
| 100 |
+
"eval_loss": 1.2222816944122314,
|
| 101 |
+
"eval_runtime": 63.8482,
|
| 102 |
+
"eval_samples_per_second": 16.414,
|
| 103 |
+
"eval_steps_per_second": 2.052,
|
| 104 |
+
"eval_wer": 0.6261278932914869,
|
| 105 |
+
"step": 1200
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.9598308668076108,
|
| 109 |
+
"grad_norm": 0.8811041116714478,
|
| 110 |
+
"learning_rate": 0.00028032140248356464,
|
| 111 |
+
"loss": 0.1382,
|
| 112 |
+
"step": 1400
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.9598308668076108,
|
| 116 |
+
"eval_loss": 1.3115968704223633,
|
| 117 |
+
"eval_runtime": 65.0526,
|
| 118 |
+
"eval_samples_per_second": 16.11,
|
| 119 |
+
"eval_steps_per_second": 2.014,
|
| 120 |
+
"eval_wer": 0.6506473126716359,
|
| 121 |
+
"step": 1400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.382663847780127,
|
| 125 |
+
"grad_norm": 0.7833051085472107,
|
| 126 |
+
"learning_rate": 0.0002759386413440467,
|
| 127 |
+
"loss": 0.1239,
|
| 128 |
+
"step": 1600
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.382663847780127,
|
| 132 |
+
"eval_loss": 1.1977170705795288,
|
| 133 |
+
"eval_runtime": 64.8745,
|
| 134 |
+
"eval_samples_per_second": 16.154,
|
| 135 |
+
"eval_steps_per_second": 2.019,
|
| 136 |
+
"eval_wer": 0.6188701451549627,
|
| 137 |
+
"step": 1600
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 3.8054968287526427,
|
| 141 |
+
"grad_norm": 0.4267895817756653,
|
| 142 |
+
"learning_rate": 0.0002715558802045288,
|
| 143 |
+
"loss": 0.1228,
|
| 144 |
+
"step": 1800
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 3.8054968287526427,
|
| 148 |
+
"eval_loss": 1.18520188331604,
|
| 149 |
+
"eval_runtime": 64.2457,
|
| 150 |
+
"eval_samples_per_second": 16.312,
|
| 151 |
+
"eval_steps_per_second": 2.039,
|
| 152 |
+
"eval_wer": 0.6280894468418988,
|
| 153 |
+
"step": 1800
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.2283298097251585,
|
| 157 |
+
"grad_norm": 0.7231135964393616,
|
| 158 |
+
"learning_rate": 0.00026717311906501094,
|
| 159 |
+
"loss": 0.1117,
|
| 160 |
+
"step": 2000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.2283298097251585,
|
| 164 |
+
"eval_loss": 1.3370016813278198,
|
| 165 |
+
"eval_runtime": 64.7812,
|
| 166 |
+
"eval_samples_per_second": 16.178,
|
| 167 |
+
"eval_steps_per_second": 2.022,
|
| 168 |
+
"eval_wer": 0.6494703805413887,
|
| 169 |
+
"step": 2000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.651162790697675,
|
| 173 |
+
"grad_norm": 0.40635955333709717,
|
| 174 |
+
"learning_rate": 0.000262790357925493,
|
| 175 |
+
"loss": 0.1118,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 4.651162790697675,
|
| 180 |
+
"eval_loss": 1.3265222311019897,
|
| 181 |
+
"eval_runtime": 63.5949,
|
| 182 |
+
"eval_samples_per_second": 16.479,
|
| 183 |
+
"eval_steps_per_second": 2.06,
|
| 184 |
+
"eval_wer": 0.6431934091800706,
|
| 185 |
+
"step": 2200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.07399577167019,
|
| 189 |
+
"grad_norm": 0.9417168498039246,
|
| 190 |
+
"learning_rate": 0.0002584075967859751,
|
| 191 |
+
"loss": 0.1101,
|
| 192 |
+
"step": 2400
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.07399577167019,
|
| 196 |
+
"eval_loss": 1.345849633216858,
|
| 197 |
+
"eval_runtime": 64.4165,
|
| 198 |
+
"eval_samples_per_second": 16.269,
|
| 199 |
+
"eval_steps_per_second": 2.034,
|
| 200 |
+
"eval_wer": 0.6310317771675167,
|
| 201 |
+
"step": 2400
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.496828752642706,
|
| 205 |
+
"grad_norm": 0.7866037487983704,
|
| 206 |
+
"learning_rate": 0.00025402483564645724,
|
| 207 |
+
"loss": 0.1328,
|
| 208 |
+
"step": 2600
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.496828752642706,
|
| 212 |
+
"eval_loss": 1.2545326948165894,
|
| 213 |
+
"eval_runtime": 64.0398,
|
| 214 |
+
"eval_samples_per_second": 16.365,
|
| 215 |
+
"eval_steps_per_second": 2.046,
|
| 216 |
+
"eval_wer": 0.6341702628481758,
|
| 217 |
+
"step": 2600
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.9196617336152215,
|
| 221 |
+
"grad_norm": 0.47020798921585083,
|
| 222 |
+
"learning_rate": 0.00024964207450693936,
|
| 223 |
+
"loss": 0.1384,
|
| 224 |
+
"step": 2800
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.9196617336152215,
|
| 228 |
+
"eval_loss": 1.2806007862091064,
|
| 229 |
+
"eval_runtime": 64.3065,
|
| 230 |
+
"eval_samples_per_second": 16.297,
|
| 231 |
+
"eval_steps_per_second": 2.037,
|
| 232 |
+
"eval_wer": 0.6265202040015693,
|
| 233 |
+
"step": 2800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.342494714587738,
|
| 237 |
+
"grad_norm": 0.605645477771759,
|
| 238 |
+
"learning_rate": 0.0002452593133674214,
|
| 239 |
+
"loss": 0.1334,
|
| 240 |
+
"step": 3000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 6.342494714587738,
|
| 244 |
+
"eval_loss": 1.2484089136123657,
|
| 245 |
+
"eval_runtime": 65.2815,
|
| 246 |
+
"eval_samples_per_second": 16.054,
|
| 247 |
+
"eval_steps_per_second": 2.007,
|
| 248 |
+
"eval_wer": 0.6369164378187524,
|
| 249 |
+
"step": 3000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 6.765327695560254,
|
| 253 |
+
"grad_norm": 0.6708455085754395,
|
| 254 |
+
"learning_rate": 0.00024087655222790357,
|
| 255 |
+
"loss": 0.1383,
|
| 256 |
+
"step": 3200
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 6.765327695560254,
|
| 260 |
+
"eval_loss": 1.2701318264007568,
|
| 261 |
+
"eval_runtime": 63.6121,
|
| 262 |
+
"eval_samples_per_second": 16.475,
|
| 263 |
+
"eval_steps_per_second": 2.059,
|
| 264 |
+
"eval_wer": 0.6479011377010593,
|
| 265 |
+
"step": 3200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 7.188160676532769,
|
| 269 |
+
"grad_norm": 0.3204992711544037,
|
| 270 |
+
"learning_rate": 0.00023649379108838566,
|
| 271 |
+
"loss": 0.1281,
|
| 272 |
+
"step": 3400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 7.188160676532769,
|
| 276 |
+
"eval_loss": 1.1926395893096924,
|
| 277 |
+
"eval_runtime": 65.1953,
|
| 278 |
+
"eval_samples_per_second": 16.075,
|
| 279 |
+
"eval_steps_per_second": 2.009,
|
| 280 |
+
"eval_wer": 0.6314240878775991,
|
| 281 |
+
"step": 3400
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 7.6109936575052854,
|
| 285 |
+
"grad_norm": 2.2028682231903076,
|
| 286 |
+
"learning_rate": 0.00023211102994886775,
|
| 287 |
+
"loss": 0.1232,
|
| 288 |
+
"step": 3600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 7.6109936575052854,
|
| 292 |
+
"eval_loss": 1.2255371809005737,
|
| 293 |
+
"eval_runtime": 65.3961,
|
| 294 |
+
"eval_samples_per_second": 16.025,
|
| 295 |
+
"eval_steps_per_second": 2.003,
|
| 296 |
+
"eval_wer": 0.6186739897999215,
|
| 297 |
+
"step": 3600
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.033826638477802,
|
| 301 |
+
"grad_norm": 0.33132538199424744,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.0727,
|
| 304 |
+
"step": 3800
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 8.033826638477802,
|
| 308 |
+
"eval_loss": 1.23982572555542,
|
| 309 |
+
"eval_runtime": 66.6836,
|
| 310 |
+
"eval_samples_per_second": 15.716,
|
| 311 |
+
"eval_steps_per_second": 1.575,
|
| 312 |
+
"eval_wer": 0.6014123185562966,
|
| 313 |
+
"step": 3800
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 8.456659619450317,
|
| 317 |
+
"grad_norm": 0.3985452950000763,
|
| 318 |
+
"learning_rate": 4e-05,
|
| 319 |
+
"loss": 0.0749,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 8.456659619450317,
|
| 324 |
+
"eval_loss": 1.2319424152374268,
|
| 325 |
+
"eval_runtime": 67.0281,
|
| 326 |
+
"eval_samples_per_second": 15.635,
|
| 327 |
+
"eval_steps_per_second": 1.567,
|
| 328 |
+
"eval_wer": 0.595723813260102,
|
| 329 |
+
"step": 4000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 8.879492600422832,
|
| 333 |
+
"grad_norm": 0.5187695622444153,
|
| 334 |
+
"learning_rate": 4.955849889624724e-05,
|
| 335 |
+
"loss": 0.0734,
|
| 336 |
+
"step": 4200
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 8.879492600422832,
|
| 340 |
+
"eval_loss": 1.2246508598327637,
|
| 341 |
+
"eval_runtime": 66.6194,
|
| 342 |
+
"eval_samples_per_second": 15.731,
|
| 343 |
+
"eval_steps_per_second": 1.576,
|
| 344 |
+
"eval_wer": 0.5878775990584543,
|
| 345 |
+
"step": 4200
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 9.30232558139535,
|
| 349 |
+
"grad_norm": 0.3908683955669403,
|
| 350 |
+
"learning_rate": 4.867549668874172e-05,
|
| 351 |
+
"loss": 0.0684,
|
| 352 |
+
"step": 4400
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 9.30232558139535,
|
| 356 |
+
"eval_loss": 1.347394585609436,
|
| 357 |
+
"eval_runtime": 65.574,
|
| 358 |
+
"eval_samples_per_second": 15.982,
|
| 359 |
+
"eval_steps_per_second": 1.601,
|
| 360 |
+
"eval_wer": 0.6135739505688506,
|
| 361 |
+
"step": 4400
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 9.725158562367865,
|
| 365 |
+
"grad_norm": 1.09392511844635,
|
| 366 |
+
"learning_rate": 4.779690949227373e-05,
|
| 367 |
+
"loss": 0.073,
|
| 368 |
+
"step": 4600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 9.725158562367865,
|
| 372 |
+
"eval_loss": 1.2836501598358154,
|
| 373 |
+
"eval_runtime": 76.5173,
|
| 374 |
+
"eval_samples_per_second": 13.696,
|
| 375 |
+
"eval_steps_per_second": 1.372,
|
| 376 |
+
"eval_wer": 0.5935661043546488,
|
| 377 |
+
"step": 4600
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 10.14799154334038,
|
| 381 |
+
"grad_norm": 0.786432683467865,
|
| 382 |
+
"learning_rate": 4.691390728476822e-05,
|
| 383 |
+
"loss": 0.0728,
|
| 384 |
+
"step": 4800
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 10.14799154334038,
|
| 388 |
+
"eval_loss": 1.247693657875061,
|
| 389 |
+
"eval_runtime": 66.6059,
|
| 390 |
+
"eval_samples_per_second": 15.734,
|
| 391 |
+
"eval_steps_per_second": 1.576,
|
| 392 |
+
"eval_wer": 0.5910160847391134,
|
| 393 |
+
"step": 4800
|
| 394 |
+
}
|
| 395 |
+
],
|
| 396 |
+
"logging_steps": 200,
|
| 397 |
+
"max_steps": 11825,
|
| 398 |
+
"num_input_tokens_seen": 0,
|
| 399 |
+
"num_train_epochs": 25,
|
| 400 |
+
"save_steps": 200,
|
| 401 |
+
"stateful_callbacks": {
|
| 402 |
+
"TrainerControl": {
|
| 403 |
+
"args": {
|
| 404 |
+
"should_epoch_stop": false,
|
| 405 |
+
"should_evaluate": false,
|
| 406 |
+
"should_log": false,
|
| 407 |
+
"should_save": true,
|
| 408 |
+
"should_training_stop": false
|
| 409 |
+
},
|
| 410 |
+
"attributes": {}
|
| 411 |
+
}
|
| 412 |
+
},
|
| 413 |
+
"total_flos": 1.5361215459079565e+19,
|
| 414 |
+
"train_batch_size": 8,
|
| 415 |
+
"trial_name": null,
|
| 416 |
+
"trial_params": null
|
| 417 |
+
}
|
checkpoint-4800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1cdb4afb19ad76c1497ce9c323bae80a0da0c233e1e1189c8502c0632a60f4e
|
| 3 |
+
size 5112
|