Commit
•
abb4d8d
1
Parent(s):
898c094
Training in progress, step 15400, checkpoint
Browse files- checkpoint-15400/config.json +109 -0
- checkpoint-15400/model.safetensors +3 -0
- checkpoint-15400/optimizer.pt +3 -0
- checkpoint-15400/preprocessor_config.json +9 -0
- checkpoint-15400/rng_state.pth +3 -0
- checkpoint-15400/scheduler.pt +3 -0
- checkpoint-15400/trainer_state.json +1265 -0
- checkpoint-15400/training_args.bin +3 -0
checkpoint-15400/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-15400/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fc3fe12c3d6f775d9e633a4d50d767d3858432c3ea4d0fd731bef291dee840e
|
| 3 |
+
size 1262180580
|
checkpoint-15400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6bb438e56691317a522c6ab8300353ac124a01fdd98aacf635a870d05d270895
|
| 3 |
+
size 2490905910
|
checkpoint-15400/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-15400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68abdf3c6161cc4f7def02a2452c53d4db454fdcef524fcc08fdbfbc00369b36
|
| 3 |
+
size 14308
|
checkpoint-15400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5fa96412c82b6978b8fa60e2816cd9960233339c4a4bccb13d11c7dc66d0ffa
|
| 3 |
+
size 1064
|
checkpoint-15400/trainer_state.json
ADDED
|
@@ -0,0 +1,1265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.18520188331604,
|
| 3 |
+
"best_model_checkpoint": "batoula187/wav2vec2-xlsr-arabic2/checkpoint-1800",
|
| 4 |
+
"epoch": 21.719454887218046,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 15400,
|
| 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 |
+
"epoch": 10.570824524312897,
|
| 397 |
+
"grad_norm": 0.22806741297245026,
|
| 398 |
+
"learning_rate": 4.6030905077262693e-05,
|
| 399 |
+
"loss": 0.0718,
|
| 400 |
+
"step": 5000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 10.570824524312897,
|
| 404 |
+
"eval_loss": 1.2471730709075928,
|
| 405 |
+
"eval_runtime": 67.4477,
|
| 406 |
+
"eval_samples_per_second": 15.538,
|
| 407 |
+
"eval_steps_per_second": 1.557,
|
| 408 |
+
"eval_wer": 0.5867006669282071,
|
| 409 |
+
"step": 5000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 10.993657505285412,
|
| 413 |
+
"grad_norm": 1.541914463043213,
|
| 414 |
+
"learning_rate": 4.5147902869757175e-05,
|
| 415 |
+
"loss": 0.0685,
|
| 416 |
+
"step": 5200
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 10.993657505285412,
|
| 420 |
+
"eval_loss": 1.2693225145339966,
|
| 421 |
+
"eval_runtime": 65.4788,
|
| 422 |
+
"eval_samples_per_second": 16.005,
|
| 423 |
+
"eval_steps_per_second": 1.604,
|
| 424 |
+
"eval_wer": 0.5788544527265594,
|
| 425 |
+
"step": 5200
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 11.416490486257928,
|
| 429 |
+
"grad_norm": 0.4897485673427582,
|
| 430 |
+
"learning_rate": 4.4264900662251656e-05,
|
| 431 |
+
"loss": 0.0649,
|
| 432 |
+
"step": 5400
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 11.416490486257928,
|
| 436 |
+
"eval_loss": 1.2164980173110962,
|
| 437 |
+
"eval_runtime": 66.1024,
|
| 438 |
+
"eval_samples_per_second": 15.854,
|
| 439 |
+
"eval_steps_per_second": 1.588,
|
| 440 |
+
"eval_wer": 0.5786582973715182,
|
| 441 |
+
"step": 5400
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 11.839323467230443,
|
| 445 |
+
"grad_norm": 0.269406795501709,
|
| 446 |
+
"learning_rate": 4.338189845474614e-05,
|
| 447 |
+
"loss": 0.0632,
|
| 448 |
+
"step": 5600
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 11.839323467230443,
|
| 452 |
+
"eval_loss": 1.2446550130844116,
|
| 453 |
+
"eval_runtime": 69.6183,
|
| 454 |
+
"eval_samples_per_second": 15.054,
|
| 455 |
+
"eval_steps_per_second": 1.508,
|
| 456 |
+
"eval_wer": 0.5841506473126716,
|
| 457 |
+
"step": 5600
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 12.26215644820296,
|
| 461 |
+
"grad_norm": 0.6048020720481873,
|
| 462 |
+
"learning_rate": 4.249889624724062e-05,
|
| 463 |
+
"loss": 0.0625,
|
| 464 |
+
"step": 5800
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 12.26215644820296,
|
| 468 |
+
"eval_loss": 1.308754324913025,
|
| 469 |
+
"eval_runtime": 67.8481,
|
| 470 |
+
"eval_samples_per_second": 15.446,
|
| 471 |
+
"eval_steps_per_second": 1.548,
|
| 472 |
+
"eval_wer": 0.5806198509219301,
|
| 473 |
+
"step": 5800
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 12.684989429175475,
|
| 477 |
+
"grad_norm": 0.2880701720714569,
|
| 478 |
+
"learning_rate": 4.16158940397351e-05,
|
| 479 |
+
"loss": 0.061,
|
| 480 |
+
"step": 6000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 12.684989429175475,
|
| 484 |
+
"eval_loss": 1.3398616313934326,
|
| 485 |
+
"eval_runtime": 67.8965,
|
| 486 |
+
"eval_samples_per_second": 15.435,
|
| 487 |
+
"eval_steps_per_second": 1.546,
|
| 488 |
+
"eval_wer": 0.5923891722244017,
|
| 489 |
+
"step": 6000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 13.10782241014799,
|
| 493 |
+
"grad_norm": 0.419969767332077,
|
| 494 |
+
"learning_rate": 4.073289183222958e-05,
|
| 495 |
+
"loss": 0.0595,
|
| 496 |
+
"step": 6200
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 13.10782241014799,
|
| 500 |
+
"eval_loss": 1.3048608303070068,
|
| 501 |
+
"eval_runtime": 66.7124,
|
| 502 |
+
"eval_samples_per_second": 15.709,
|
| 503 |
+
"eval_steps_per_second": 1.574,
|
| 504 |
+
"eval_wer": 0.5768928991761475,
|
| 505 |
+
"step": 6200
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 13.530655391120508,
|
| 509 |
+
"grad_norm": 0.17939791083335876,
|
| 510 |
+
"learning_rate": 3.9849889624724064e-05,
|
| 511 |
+
"loss": 0.0608,
|
| 512 |
+
"step": 6400
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 13.530655391120508,
|
| 516 |
+
"eval_loss": 1.2737226486206055,
|
| 517 |
+
"eval_runtime": 67.084,
|
| 518 |
+
"eval_samples_per_second": 15.622,
|
| 519 |
+
"eval_steps_per_second": 1.565,
|
| 520 |
+
"eval_wer": 0.573362102785406,
|
| 521 |
+
"step": 6400
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 13.953488372093023,
|
| 525 |
+
"grad_norm": 0.32987338304519653,
|
| 526 |
+
"learning_rate": 3.8966887417218545e-05,
|
| 527 |
+
"loss": 0.0596,
|
| 528 |
+
"step": 6600
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 13.953488372093023,
|
| 532 |
+
"eval_loss": 1.228838324546814,
|
| 533 |
+
"eval_runtime": 66.3842,
|
| 534 |
+
"eval_samples_per_second": 15.787,
|
| 535 |
+
"eval_steps_per_second": 1.582,
|
| 536 |
+
"eval_wer": 0.5747351902706944,
|
| 537 |
+
"step": 6600
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 14.376321353065538,
|
| 541 |
+
"grad_norm": 1.2768590450286865,
|
| 542 |
+
"learning_rate": 3.808388520971303e-05,
|
| 543 |
+
"loss": 0.0565,
|
| 544 |
+
"step": 6800
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 14.376321353065538,
|
| 548 |
+
"eval_loss": 1.2599496841430664,
|
| 549 |
+
"eval_runtime": 66.2771,
|
| 550 |
+
"eval_samples_per_second": 15.812,
|
| 551 |
+
"eval_steps_per_second": 1.584,
|
| 552 |
+
"eval_wer": 0.5676735974892114,
|
| 553 |
+
"step": 6800
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 14.799154334038056,
|
| 557 |
+
"grad_norm": 0.410118043422699,
|
| 558 |
+
"learning_rate": 3.720088300220751e-05,
|
| 559 |
+
"loss": 0.0568,
|
| 560 |
+
"step": 7000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 14.799154334038056,
|
| 564 |
+
"eval_loss": 1.2704553604125977,
|
| 565 |
+
"eval_runtime": 66.0979,
|
| 566 |
+
"eval_samples_per_second": 15.855,
|
| 567 |
+
"eval_steps_per_second": 1.589,
|
| 568 |
+
"eval_wer": 0.5621812475480581,
|
| 569 |
+
"step": 7000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 15.221987315010571,
|
| 573 |
+
"grad_norm": 0.3582391142845154,
|
| 574 |
+
"learning_rate": 3.631788079470198e-05,
|
| 575 |
+
"loss": 0.0538,
|
| 576 |
+
"step": 7200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 15.221987315010571,
|
| 580 |
+
"eval_loss": 1.3540122509002686,
|
| 581 |
+
"eval_runtime": 66.4126,
|
| 582 |
+
"eval_samples_per_second": 15.78,
|
| 583 |
+
"eval_steps_per_second": 1.581,
|
| 584 |
+
"eval_wer": 0.5837583366025892,
|
| 585 |
+
"step": 7200
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 15.644820295983086,
|
| 589 |
+
"grad_norm": 0.261311411857605,
|
| 590 |
+
"learning_rate": 3.5439293598234e-05,
|
| 591 |
+
"loss": 0.0585,
|
| 592 |
+
"step": 7400
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 15.644820295983086,
|
| 596 |
+
"eval_loss": 1.3333740234375,
|
| 597 |
+
"eval_runtime": 66.9559,
|
| 598 |
+
"eval_samples_per_second": 15.652,
|
| 599 |
+
"eval_steps_per_second": 1.568,
|
| 600 |
+
"eval_wer": 0.5798352295017654,
|
| 601 |
+
"step": 7400
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 16.067653276955603,
|
| 605 |
+
"grad_norm": 0.3650034964084625,
|
| 606 |
+
"learning_rate": 3.455629139072848e-05,
|
| 607 |
+
"loss": 0.0548,
|
| 608 |
+
"step": 7600
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 16.067653276955603,
|
| 612 |
+
"eval_loss": 1.3312934637069702,
|
| 613 |
+
"eval_runtime": 66.7147,
|
| 614 |
+
"eval_samples_per_second": 15.709,
|
| 615 |
+
"eval_steps_per_second": 1.574,
|
| 616 |
+
"eval_wer": 0.5723813260102001,
|
| 617 |
+
"step": 7600
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 16.49048625792812,
|
| 621 |
+
"grad_norm": 0.4263480305671692,
|
| 622 |
+
"learning_rate": 3.367328918322296e-05,
|
| 623 |
+
"loss": 0.0526,
|
| 624 |
+
"step": 7800
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 16.49048625792812,
|
| 628 |
+
"eval_loss": 1.3299189805984497,
|
| 629 |
+
"eval_runtime": 67.2527,
|
| 630 |
+
"eval_samples_per_second": 15.583,
|
| 631 |
+
"eval_steps_per_second": 1.561,
|
| 632 |
+
"eval_wer": 0.5719890153001177,
|
| 633 |
+
"step": 7800
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 16.913319238900634,
|
| 637 |
+
"grad_norm": 0.41519472002983093,
|
| 638 |
+
"learning_rate": 3.2790286975717444e-05,
|
| 639 |
+
"loss": 0.0577,
|
| 640 |
+
"step": 8000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 16.913319238900634,
|
| 644 |
+
"eval_loss": 1.3206344842910767,
|
| 645 |
+
"eval_runtime": 67.3412,
|
| 646 |
+
"eval_samples_per_second": 15.563,
|
| 647 |
+
"eval_steps_per_second": 1.559,
|
| 648 |
+
"eval_wer": 0.5829737151824245,
|
| 649 |
+
"step": 8000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 17.33615221987315,
|
| 653 |
+
"grad_norm": 0.9854322075843811,
|
| 654 |
+
"learning_rate": 3.190728476821192e-05,
|
| 655 |
+
"loss": 0.0513,
|
| 656 |
+
"step": 8200
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 17.33615221987315,
|
| 660 |
+
"eval_loss": 1.3500277996063232,
|
| 661 |
+
"eval_runtime": 66.908,
|
| 662 |
+
"eval_samples_per_second": 15.663,
|
| 663 |
+
"eval_steps_per_second": 1.569,
|
| 664 |
+
"eval_wer": 0.5786582973715182,
|
| 665 |
+
"step": 8200
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 17.758985200845665,
|
| 669 |
+
"grad_norm": 0.4509384334087372,
|
| 670 |
+
"learning_rate": 3.10242825607064e-05,
|
| 671 |
+
"loss": 0.0506,
|
| 672 |
+
"step": 8400
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 17.758985200845665,
|
| 676 |
+
"eval_loss": 1.3184752464294434,
|
| 677 |
+
"eval_runtime": 66.0674,
|
| 678 |
+
"eval_samples_per_second": 15.863,
|
| 679 |
+
"eval_steps_per_second": 1.589,
|
| 680 |
+
"eval_wer": 0.5698313063946646,
|
| 681 |
+
"step": 8400
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 18.181818181818183,
|
| 685 |
+
"grad_norm": 0.20881137251853943,
|
| 686 |
+
"learning_rate": 3.0145695364238412e-05,
|
| 687 |
+
"loss": 0.0498,
|
| 688 |
+
"step": 8600
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 18.181818181818183,
|
| 692 |
+
"eval_loss": 1.3655884265899658,
|
| 693 |
+
"eval_runtime": 65.7835,
|
| 694 |
+
"eval_samples_per_second": 15.931,
|
| 695 |
+
"eval_steps_per_second": 1.596,
|
| 696 |
+
"eval_wer": 0.5800313848568066,
|
| 697 |
+
"step": 8600
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 18.6046511627907,
|
| 701 |
+
"grad_norm": 8.674209594726562,
|
| 702 |
+
"learning_rate": 2.9262693156732894e-05,
|
| 703 |
+
"loss": 0.0515,
|
| 704 |
+
"step": 8800
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 18.6046511627907,
|
| 708 |
+
"eval_loss": 1.3253123760223389,
|
| 709 |
+
"eval_runtime": 65.9715,
|
| 710 |
+
"eval_samples_per_second": 15.886,
|
| 711 |
+
"eval_steps_per_second": 1.592,
|
| 712 |
+
"eval_wer": 0.5668889760690466,
|
| 713 |
+
"step": 8800
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 19.027484143763214,
|
| 717 |
+
"grad_norm": 0.2627590298652649,
|
| 718 |
+
"learning_rate": 2.83841059602649e-05,
|
| 719 |
+
"loss": 0.05,
|
| 720 |
+
"step": 9000
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 19.027484143763214,
|
| 724 |
+
"eval_loss": 1.3410530090332031,
|
| 725 |
+
"eval_runtime": 65.3408,
|
| 726 |
+
"eval_samples_per_second": 16.039,
|
| 727 |
+
"eval_steps_per_second": 1.607,
|
| 728 |
+
"eval_wer": 0.5810121616320125,
|
| 729 |
+
"step": 9000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 19.45031712473573,
|
| 733 |
+
"grad_norm": 0.3467521667480469,
|
| 734 |
+
"learning_rate": 2.7501103752759385e-05,
|
| 735 |
+
"loss": 0.048,
|
| 736 |
+
"step": 9200
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 19.45031712473573,
|
| 740 |
+
"eval_loss": 1.3627517223358154,
|
| 741 |
+
"eval_runtime": 67.7618,
|
| 742 |
+
"eval_samples_per_second": 15.466,
|
| 743 |
+
"eval_steps_per_second": 1.55,
|
| 744 |
+
"eval_wer": 0.5729697920753236,
|
| 745 |
+
"step": 9200
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 19.873150105708245,
|
| 749 |
+
"grad_norm": 0.3944872319698334,
|
| 750 |
+
"learning_rate": 2.6618101545253866e-05,
|
| 751 |
+
"loss": 0.049,
|
| 752 |
+
"step": 9400
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 19.873150105708245,
|
| 756 |
+
"eval_loss": 1.3700072765350342,
|
| 757 |
+
"eval_runtime": 71.6559,
|
| 758 |
+
"eval_samples_per_second": 14.625,
|
| 759 |
+
"eval_steps_per_second": 1.465,
|
| 760 |
+
"eval_wer": 0.5729697920753236,
|
| 761 |
+
"step": 9400
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 20.29598308668076,
|
| 765 |
+
"grad_norm": 0.63824862241745,
|
| 766 |
+
"learning_rate": 2.5735099337748348e-05,
|
| 767 |
+
"loss": 0.0469,
|
| 768 |
+
"step": 9600
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 20.29598308668076,
|
| 772 |
+
"eval_loss": 1.364588975906372,
|
| 773 |
+
"eval_runtime": 75.7846,
|
| 774 |
+
"eval_samples_per_second": 13.829,
|
| 775 |
+
"eval_steps_per_second": 1.386,
|
| 776 |
+
"eval_wer": 0.5717928599450764,
|
| 777 |
+
"step": 9600
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 20.718816067653275,
|
| 781 |
+
"grad_norm": 0.29639732837677,
|
| 782 |
+
"learning_rate": 2.4852097130242826e-05,
|
| 783 |
+
"loss": 0.0474,
|
| 784 |
+
"step": 9800
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 20.718816067653275,
|
| 788 |
+
"eval_loss": 1.4190597534179688,
|
| 789 |
+
"eval_runtime": 72.4218,
|
| 790 |
+
"eval_samples_per_second": 14.471,
|
| 791 |
+
"eval_steps_per_second": 1.45,
|
| 792 |
+
"eval_wer": 0.5786582973715182,
|
| 793 |
+
"step": 9800
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 21.141649048625794,
|
| 797 |
+
"grad_norm": 0.3832600712776184,
|
| 798 |
+
"learning_rate": 2.396909492273731e-05,
|
| 799 |
+
"loss": 0.0488,
|
| 800 |
+
"step": 10000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 21.141649048625794,
|
| 804 |
+
"eval_loss": 1.3449620008468628,
|
| 805 |
+
"eval_runtime": 76.5571,
|
| 806 |
+
"eval_samples_per_second": 13.689,
|
| 807 |
+
"eval_steps_per_second": 1.372,
|
| 808 |
+
"eval_wer": 0.575323656335818,
|
| 809 |
+
"step": 10000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 21.56448202959831,
|
| 813 |
+
"grad_norm": 0.44069257378578186,
|
| 814 |
+
"learning_rate": 2.308609271523179e-05,
|
| 815 |
+
"loss": 0.0466,
|
| 816 |
+
"step": 10200
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 21.56448202959831,
|
| 820 |
+
"eval_loss": 1.2960785627365112,
|
| 821 |
+
"eval_runtime": 72.0544,
|
| 822 |
+
"eval_samples_per_second": 14.545,
|
| 823 |
+
"eval_steps_per_second": 1.457,
|
| 824 |
+
"eval_wer": 0.5612004707728521,
|
| 825 |
+
"step": 10200
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 21.987315010570825,
|
| 829 |
+
"grad_norm": 0.336251437664032,
|
| 830 |
+
"learning_rate": 2.220309050772627e-05,
|
| 831 |
+
"loss": 0.0462,
|
| 832 |
+
"step": 10400
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 21.987315010570825,
|
| 836 |
+
"eval_loss": 1.3378616571426392,
|
| 837 |
+
"eval_runtime": 66.2158,
|
| 838 |
+
"eval_samples_per_second": 15.827,
|
| 839 |
+
"eval_steps_per_second": 1.586,
|
| 840 |
+
"eval_wer": 0.5731659474303649,
|
| 841 |
+
"step": 10400
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 22.41014799154334,
|
| 845 |
+
"grad_norm": 0.24060240387916565,
|
| 846 |
+
"learning_rate": 2.1320088300220752e-05,
|
| 847 |
+
"loss": 0.0479,
|
| 848 |
+
"step": 10600
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 22.41014799154334,
|
| 852 |
+
"eval_loss": 1.364053726196289,
|
| 853 |
+
"eval_runtime": 65.0016,
|
| 854 |
+
"eval_samples_per_second": 16.123,
|
| 855 |
+
"eval_steps_per_second": 1.615,
|
| 856 |
+
"eval_wer": 0.5755198116908592,
|
| 857 |
+
"step": 10600
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 22.832980972515855,
|
| 861 |
+
"grad_norm": 0.28006625175476074,
|
| 862 |
+
"learning_rate": 2.0437086092715233e-05,
|
| 863 |
+
"loss": 0.0475,
|
| 864 |
+
"step": 10800
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 22.832980972515855,
|
| 868 |
+
"eval_loss": 1.3316149711608887,
|
| 869 |
+
"eval_runtime": 66.7406,
|
| 870 |
+
"eval_samples_per_second": 15.703,
|
| 871 |
+
"eval_steps_per_second": 1.573,
|
| 872 |
+
"eval_wer": 0.5751275009807768,
|
| 873 |
+
"step": 10800
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 23.25581395348837,
|
| 877 |
+
"grad_norm": 1.4817793369293213,
|
| 878 |
+
"learning_rate": 1.9554083885209715e-05,
|
| 879 |
+
"loss": 0.0461,
|
| 880 |
+
"step": 11000
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 23.25581395348837,
|
| 884 |
+
"eval_loss": 1.4020766019821167,
|
| 885 |
+
"eval_runtime": 65.2476,
|
| 886 |
+
"eval_samples_per_second": 16.062,
|
| 887 |
+
"eval_steps_per_second": 1.609,
|
| 888 |
+
"eval_wer": 0.5778736759513535,
|
| 889 |
+
"step": 11000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 23.67864693446089,
|
| 893 |
+
"grad_norm": 0.13971921801567078,
|
| 894 |
+
"learning_rate": 1.8671081677704193e-05,
|
| 895 |
+
"loss": 0.0443,
|
| 896 |
+
"step": 11200
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 23.67864693446089,
|
| 900 |
+
"eval_loss": 1.3807997703552246,
|
| 901 |
+
"eval_runtime": 65.7311,
|
| 902 |
+
"eval_samples_per_second": 15.944,
|
| 903 |
+
"eval_steps_per_second": 1.597,
|
| 904 |
+
"eval_wer": 0.5766967438211064,
|
| 905 |
+
"step": 11200
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 24.101479915433405,
|
| 909 |
+
"grad_norm": 1.3055016994476318,
|
| 910 |
+
"learning_rate": 1.7788079470198678e-05,
|
| 911 |
+
"loss": 0.0448,
|
| 912 |
+
"step": 11400
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 24.101479915433405,
|
| 916 |
+
"eval_loss": 1.4156945943832397,
|
| 917 |
+
"eval_runtime": 65.6108,
|
| 918 |
+
"eval_samples_per_second": 15.973,
|
| 919 |
+
"eval_steps_per_second": 1.6,
|
| 920 |
+
"eval_wer": 0.5778736759513535,
|
| 921 |
+
"step": 11400
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 16.360902255639097,
|
| 925 |
+
"grad_norm": 0.8831915259361267,
|
| 926 |
+
"learning_rate": 2.8240928882438317e-05,
|
| 927 |
+
"loss": 0.1948,
|
| 928 |
+
"step": 11600
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 16.360902255639097,
|
| 932 |
+
"eval_loss": 0.8630273342132568,
|
| 933 |
+
"eval_runtime": 66.0147,
|
| 934 |
+
"eval_samples_per_second": 15.875,
|
| 935 |
+
"eval_steps_per_second": 1.591,
|
| 936 |
+
"eval_wer": 0.5619850921930168,
|
| 937 |
+
"step": 11600
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 16.642857142857142,
|
| 941 |
+
"grad_norm": 3.0019664764404297,
|
| 942 |
+
"learning_rate": 2.7663280116110306e-05,
|
| 943 |
+
"loss": 0.1658,
|
| 944 |
+
"step": 11800
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 16.642857142857142,
|
| 948 |
+
"eval_loss": 0.933027982711792,
|
| 949 |
+
"eval_runtime": 65.0734,
|
| 950 |
+
"eval_samples_per_second": 16.105,
|
| 951 |
+
"eval_steps_per_second": 1.614,
|
| 952 |
+
"eval_wer": 0.569242840329541,
|
| 953 |
+
"step": 11800
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 16.924812030075188,
|
| 957 |
+
"grad_norm": 1.1889337301254272,
|
| 958 |
+
"learning_rate": 2.7082728592162554e-05,
|
| 959 |
+
"loss": 0.1632,
|
| 960 |
+
"step": 12000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 16.924812030075188,
|
| 964 |
+
"eval_loss": 0.8790073394775391,
|
| 965 |
+
"eval_runtime": 66.0534,
|
| 966 |
+
"eval_samples_per_second": 15.866,
|
| 967 |
+
"eval_steps_per_second": 1.59,
|
| 968 |
+
"eval_wer": 0.5517850137308749,
|
| 969 |
+
"step": 12000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 17.206766917293233,
|
| 973 |
+
"grad_norm": 0.7651678919792175,
|
| 974 |
+
"learning_rate": 2.6502177068214808e-05,
|
| 975 |
+
"loss": 0.1373,
|
| 976 |
+
"step": 12200
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 17.206766917293233,
|
| 980 |
+
"eval_loss": 0.9278713464736938,
|
| 981 |
+
"eval_runtime": 66.2333,
|
| 982 |
+
"eval_samples_per_second": 15.823,
|
| 983 |
+
"eval_steps_per_second": 1.585,
|
| 984 |
+
"eval_wer": 0.5455080423695566,
|
| 985 |
+
"step": 12200
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 17.48872180451128,
|
| 989 |
+
"grad_norm": 0.5034601092338562,
|
| 990 |
+
"learning_rate": 2.5921625544267052e-05,
|
| 991 |
+
"loss": 0.1233,
|
| 992 |
+
"step": 12400
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 17.48872180451128,
|
| 996 |
+
"eval_loss": 1.0113743543624878,
|
| 997 |
+
"eval_runtime": 64.1889,
|
| 998 |
+
"eval_samples_per_second": 16.327,
|
| 999 |
+
"eval_steps_per_second": 1.636,
|
| 1000 |
+
"eval_wer": 0.5633581796783053,
|
| 1001 |
+
"step": 12400
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 17.770676691729324,
|
| 1005 |
+
"grad_norm": 0.6201246380805969,
|
| 1006 |
+
"learning_rate": 2.5341074020319307e-05,
|
| 1007 |
+
"loss": 0.1223,
|
| 1008 |
+
"step": 12600
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 17.770676691729324,
|
| 1012 |
+
"eval_loss": 1.0203057527542114,
|
| 1013 |
+
"eval_runtime": 65.0796,
|
| 1014 |
+
"eval_samples_per_second": 16.103,
|
| 1015 |
+
"eval_steps_per_second": 1.613,
|
| 1016 |
+
"eval_wer": 0.5637504903883876,
|
| 1017 |
+
"step": 12600
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 18.05263157894737,
|
| 1021 |
+
"grad_norm": 0.47344139218330383,
|
| 1022 |
+
"learning_rate": 2.4760522496371554e-05,
|
| 1023 |
+
"loss": 0.1207,
|
| 1024 |
+
"step": 12800
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 18.05263157894737,
|
| 1028 |
+
"eval_loss": 1.0659551620483398,
|
| 1029 |
+
"eval_runtime": 65.8453,
|
| 1030 |
+
"eval_samples_per_second": 15.916,
|
| 1031 |
+
"eval_steps_per_second": 1.595,
|
| 1032 |
+
"eval_wer": 0.5723813260102001,
|
| 1033 |
+
"step": 12800
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 18.334586466165412,
|
| 1037 |
+
"grad_norm": 0.9515664577484131,
|
| 1038 |
+
"learning_rate": 2.4179970972423805e-05,
|
| 1039 |
+
"loss": 0.1009,
|
| 1040 |
+
"step": 13000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 18.334586466165412,
|
| 1044 |
+
"eval_loss": 1.0872668027877808,
|
| 1045 |
+
"eval_runtime": 66.957,
|
| 1046 |
+
"eval_samples_per_second": 15.652,
|
| 1047 |
+
"eval_steps_per_second": 1.568,
|
| 1048 |
+
"eval_wer": 0.5666928207140055,
|
| 1049 |
+
"step": 13000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 18.616541353383457,
|
| 1053 |
+
"grad_norm": 2.7835988998413086,
|
| 1054 |
+
"learning_rate": 2.3599419448476053e-05,
|
| 1055 |
+
"loss": 0.106,
|
| 1056 |
+
"step": 13200
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 18.616541353383457,
|
| 1060 |
+
"eval_loss": 1.1188278198242188,
|
| 1061 |
+
"eval_runtime": 65.7163,
|
| 1062 |
+
"eval_samples_per_second": 15.947,
|
| 1063 |
+
"eval_steps_per_second": 1.598,
|
| 1064 |
+
"eval_wer": 0.5666928207140055,
|
| 1065 |
+
"step": 13200
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 18.898496240601503,
|
| 1069 |
+
"grad_norm": 0.33580583333969116,
|
| 1070 |
+
"learning_rate": 2.3018867924528304e-05,
|
| 1071 |
+
"loss": 0.0989,
|
| 1072 |
+
"step": 13400
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 18.898496240601503,
|
| 1076 |
+
"eval_loss": 1.0954115390777588,
|
| 1077 |
+
"eval_runtime": 67.9533,
|
| 1078 |
+
"eval_samples_per_second": 15.422,
|
| 1079 |
+
"eval_steps_per_second": 1.545,
|
| 1080 |
+
"eval_wer": 0.5688505296194586,
|
| 1081 |
+
"step": 13400
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 19.18045112781955,
|
| 1085 |
+
"grad_norm": 0.20191040635108948,
|
| 1086 |
+
"learning_rate": 2.243831640058055e-05,
|
| 1087 |
+
"loss": 0.0981,
|
| 1088 |
+
"step": 13600
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 19.18045112781955,
|
| 1092 |
+
"eval_loss": 1.1167967319488525,
|
| 1093 |
+
"eval_runtime": 68.7147,
|
| 1094 |
+
"eval_samples_per_second": 15.251,
|
| 1095 |
+
"eval_steps_per_second": 1.528,
|
| 1096 |
+
"eval_wer": 0.5635543350333464,
|
| 1097 |
+
"step": 13600
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 19.463815789473685,
|
| 1101 |
+
"grad_norm": 0.5584314465522766,
|
| 1102 |
+
"learning_rate": 2.186066763425254e-05,
|
| 1103 |
+
"loss": 0.0858,
|
| 1104 |
+
"step": 13800
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 19.463815789473685,
|
| 1108 |
+
"eval_loss": 1.1654815673828125,
|
| 1109 |
+
"eval_runtime": 67.1626,
|
| 1110 |
+
"eval_samples_per_second": 15.604,
|
| 1111 |
+
"eval_steps_per_second": 1.563,
|
| 1112 |
+
"eval_wer": 0.5668889760690466,
|
| 1113 |
+
"step": 13800
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 19.74577067669173,
|
| 1117 |
+
"grad_norm": 0.5650402307510376,
|
| 1118 |
+
"learning_rate": 2.128011611030479e-05,
|
| 1119 |
+
"loss": 0.0851,
|
| 1120 |
+
"step": 14000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 19.74577067669173,
|
| 1124 |
+
"eval_loss": 1.1516063213348389,
|
| 1125 |
+
"eval_runtime": 66.1494,
|
| 1126 |
+
"eval_samples_per_second": 15.843,
|
| 1127 |
+
"eval_steps_per_second": 1.587,
|
| 1128 |
+
"eval_wer": 0.5596312279325225,
|
| 1129 |
+
"step": 14000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 20.027725563909776,
|
| 1133 |
+
"grad_norm": 0.35859623551368713,
|
| 1134 |
+
"learning_rate": 2.069956458635704e-05,
|
| 1135 |
+
"loss": 0.0929,
|
| 1136 |
+
"step": 14200
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 20.027725563909776,
|
| 1140 |
+
"eval_loss": 1.1067341566085815,
|
| 1141 |
+
"eval_runtime": 65.8487,
|
| 1142 |
+
"eval_samples_per_second": 15.915,
|
| 1143 |
+
"eval_steps_per_second": 1.595,
|
| 1144 |
+
"eval_wer": 0.5545311887014516,
|
| 1145 |
+
"step": 14200
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 20.309680451127818,
|
| 1149 |
+
"grad_norm": 0.4309854507446289,
|
| 1150 |
+
"learning_rate": 2.011901306240929e-05,
|
| 1151 |
+
"loss": 0.0816,
|
| 1152 |
+
"step": 14400
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 20.309680451127818,
|
| 1156 |
+
"eval_loss": 1.147910714149475,
|
| 1157 |
+
"eval_runtime": 66.4223,
|
| 1158 |
+
"eval_samples_per_second": 15.778,
|
| 1159 |
+
"eval_steps_per_second": 1.581,
|
| 1160 |
+
"eval_wer": 0.5608081600627697,
|
| 1161 |
+
"step": 14400
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 20.591635338345863,
|
| 1165 |
+
"grad_norm": 0.8902326226234436,
|
| 1166 |
+
"learning_rate": 1.9538461538461537e-05,
|
| 1167 |
+
"loss": 0.0853,
|
| 1168 |
+
"step": 14600
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 20.591635338345863,
|
| 1172 |
+
"eval_loss": 1.157361388206482,
|
| 1173 |
+
"eval_runtime": 66.7554,
|
| 1174 |
+
"eval_samples_per_second": 15.699,
|
| 1175 |
+
"eval_steps_per_second": 1.573,
|
| 1176 |
+
"eval_wer": 0.5625735582581405,
|
| 1177 |
+
"step": 14600
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 20.87359022556391,
|
| 1181 |
+
"grad_norm": 0.7075725793838501,
|
| 1182 |
+
"learning_rate": 1.8960812772133526e-05,
|
| 1183 |
+
"loss": 0.0823,
|
| 1184 |
+
"step": 14800
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 20.87359022556391,
|
| 1188 |
+
"eval_loss": 1.178560495376587,
|
| 1189 |
+
"eval_runtime": 67.6139,
|
| 1190 |
+
"eval_samples_per_second": 15.5,
|
| 1191 |
+
"eval_steps_per_second": 1.553,
|
| 1192 |
+
"eval_wer": 0.5655158885837583,
|
| 1193 |
+
"step": 14800
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 21.155545112781954,
|
| 1197 |
+
"grad_norm": 0.28010666370391846,
|
| 1198 |
+
"learning_rate": 1.8380261248185777e-05,
|
| 1199 |
+
"loss": 0.0837,
|
| 1200 |
+
"step": 15000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 21.155545112781954,
|
| 1204 |
+
"eval_loss": 1.180931568145752,
|
| 1205 |
+
"eval_runtime": 66.8333,
|
| 1206 |
+
"eval_samples_per_second": 15.681,
|
| 1207 |
+
"eval_steps_per_second": 1.571,
|
| 1208 |
+
"eval_wer": 0.5617889368379757,
|
| 1209 |
+
"step": 15000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 21.4375,
|
| 1213 |
+
"grad_norm": 0.8791319131851196,
|
| 1214 |
+
"learning_rate": 1.7799709724238025e-05,
|
| 1215 |
+
"loss": 0.0806,
|
| 1216 |
+
"step": 15200
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 21.4375,
|
| 1220 |
+
"eval_loss": 1.1776329278945923,
|
| 1221 |
+
"eval_runtime": 66.6056,
|
| 1222 |
+
"eval_samples_per_second": 15.734,
|
| 1223 |
+
"eval_steps_per_second": 1.576,
|
| 1224 |
+
"eval_wer": 0.5547273440564927,
|
| 1225 |
+
"step": 15200
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 21.719454887218046,
|
| 1229 |
+
"grad_norm": 0.5617702007293701,
|
| 1230 |
+
"learning_rate": 1.721915820029028e-05,
|
| 1231 |
+
"loss": 0.0819,
|
| 1232 |
+
"step": 15400
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 21.719454887218046,
|
| 1236 |
+
"eval_loss": 1.1667627096176147,
|
| 1237 |
+
"eval_runtime": 68.9763,
|
| 1238 |
+
"eval_samples_per_second": 15.194,
|
| 1239 |
+
"eval_steps_per_second": 1.522,
|
| 1240 |
+
"eval_wer": 0.5580619850921931,
|
| 1241 |
+
"step": 15400
|
| 1242 |
+
}
|
| 1243 |
+
],
|
| 1244 |
+
"logging_steps": 200,
|
| 1245 |
+
"max_steps": 17725,
|
| 1246 |
+
"num_input_tokens_seen": 0,
|
| 1247 |
+
"num_train_epochs": 25,
|
| 1248 |
+
"save_steps": 200,
|
| 1249 |
+
"stateful_callbacks": {
|
| 1250 |
+
"TrainerControl": {
|
| 1251 |
+
"args": {
|
| 1252 |
+
"should_epoch_stop": false,
|
| 1253 |
+
"should_evaluate": false,
|
| 1254 |
+
"should_log": false,
|
| 1255 |
+
"should_save": true,
|
| 1256 |
+
"should_training_stop": false
|
| 1257 |
+
},
|
| 1258 |
+
"attributes": {}
|
| 1259 |
+
}
|
| 1260 |
+
},
|
| 1261 |
+
"total_flos": 4.916231529580259e+19,
|
| 1262 |
+
"train_batch_size": 8,
|
| 1263 |
+
"trial_name": null,
|
| 1264 |
+
"trial_params": null
|
| 1265 |
+
}
|
checkpoint-15400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1cdb4afb19ad76c1497ce9c323bae80a0da0c233e1e1189c8502c0632a60f4e
|
| 3 |
+
size 5112
|