Upload 8 files
Browse files- config.json +82 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +11 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1185 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/w2v-bert-2.0",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_act": "relu",
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": true,
|
| 8 |
+
"apply_spec_augment": false,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2BertForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 768,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"conformer_conv_dropout": 0.1,
|
| 17 |
+
"contrastive_logits_temperature": 0.1,
|
| 18 |
+
"conv_depthwise_kernel_size": 31,
|
| 19 |
+
"ctc_loss_reduction": "mean",
|
| 20 |
+
"ctc_zero_infinity": true,
|
| 21 |
+
"diversity_loss_weight": 0.1,
|
| 22 |
+
"eos_token_id": 2,
|
| 23 |
+
"feat_proj_dropout": 0.1,
|
| 24 |
+
"feat_quantizer_dropout": 0.0,
|
| 25 |
+
"feature_projection_input_dim": 160,
|
| 26 |
+
"final_dropout": 0.1,
|
| 27 |
+
"hidden_act": "swish",
|
| 28 |
+
"hidden_dropout": 0.1,
|
| 29 |
+
"hidden_size": 1024,
|
| 30 |
+
"initializer_range": 0.02,
|
| 31 |
+
"intermediate_size": 4096,
|
| 32 |
+
"layer_norm_eps": 1e-05,
|
| 33 |
+
"layerdrop": 0.0,
|
| 34 |
+
"left_max_position_embeddings": 64,
|
| 35 |
+
"mask_feature_length": 10,
|
| 36 |
+
"mask_feature_min_masks": 0,
|
| 37 |
+
"mask_feature_prob": 0.0,
|
| 38 |
+
"mask_time_length": 10,
|
| 39 |
+
"mask_time_min_masks": 2,
|
| 40 |
+
"mask_time_prob": 0.0,
|
| 41 |
+
"max_source_positions": 5000,
|
| 42 |
+
"model_type": "wav2vec2-bert",
|
| 43 |
+
"num_adapter_layers": 1,
|
| 44 |
+
"num_attention_heads": 16,
|
| 45 |
+
"num_codevector_groups": 2,
|
| 46 |
+
"num_codevectors_per_group": 320,
|
| 47 |
+
"num_hidden_layers": 24,
|
| 48 |
+
"num_negatives": 100,
|
| 49 |
+
"output_hidden_size": 1024,
|
| 50 |
+
"pad_token_id": 67,
|
| 51 |
+
"position_embeddings_type": "relative_key",
|
| 52 |
+
"proj_codevector_dim": 768,
|
| 53 |
+
"right_max_position_embeddings": 8,
|
| 54 |
+
"rotary_embedding_base": 10000,
|
| 55 |
+
"tdnn_dilation": [
|
| 56 |
+
1,
|
| 57 |
+
2,
|
| 58 |
+
3,
|
| 59 |
+
1,
|
| 60 |
+
1
|
| 61 |
+
],
|
| 62 |
+
"tdnn_dim": [
|
| 63 |
+
512,
|
| 64 |
+
512,
|
| 65 |
+
512,
|
| 66 |
+
512,
|
| 67 |
+
1500
|
| 68 |
+
],
|
| 69 |
+
"tdnn_kernel": [
|
| 70 |
+
5,
|
| 71 |
+
3,
|
| 72 |
+
3,
|
| 73 |
+
1,
|
| 74 |
+
1
|
| 75 |
+
],
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.41.1",
|
| 78 |
+
"use_intermediate_ffn_before_adapter": false,
|
| 79 |
+
"use_weighted_layer_sum": false,
|
| 80 |
+
"vocab_size": 70,
|
| 81 |
+
"xvector_output_dim": 512
|
| 82 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:627153df92dcf757ed2427fae72bc7485db72b76e1b4ffd8415579a57c5000a8
|
| 3 |
+
size 2423101560
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77ae39e4d6fc616e8ebb66d0412f585e616cd7294c984fdbe1cd73aae06b1415
|
| 3 |
+
size 4846663914
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"feature_extractor_type": "SeamlessM4TFeatureExtractor",
|
| 3 |
+
"feature_size": 80,
|
| 4 |
+
"num_mel_bins": 80,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2BertProcessor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000,
|
| 10 |
+
"stride": 2
|
| 11 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80de60c532ddc5eb80d3b807f7b2688b1cc8537a0fea9a112532d3f9c9de7cb2
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a094e3280d2a82d7d9038b717bfcd2521b5d483a517a1b30355562f0e2dc8e7
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.4596464258262873,
|
| 5 |
+
"eval_steps": 300,
|
| 6 |
+
"global_step": 21600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.034161755914253994,
|
| 13 |
+
"grad_norm": 121.1594467163086,
|
| 14 |
+
"learning_rate": 2.97e-05,
|
| 15 |
+
"loss": 2.9674,
|
| 16 |
+
"step": 300
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.034161755914253994,
|
| 20 |
+
"eval_loss": 1.3304753303527832,
|
| 21 |
+
"eval_runtime": 352.5943,
|
| 22 |
+
"eval_samples_per_second": 10.488,
|
| 23 |
+
"eval_steps_per_second": 10.488,
|
| 24 |
+
"eval_wer": 0.7000701638532337,
|
| 25 |
+
"step": 300
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.06832351182850799,
|
| 29 |
+
"grad_norm": 23.02712631225586,
|
| 30 |
+
"learning_rate": 4.986136783733826e-05,
|
| 31 |
+
"loss": 1.2476,
|
| 32 |
+
"step": 600
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06832351182850799,
|
| 36 |
+
"eval_loss": 1.1660360097885132,
|
| 37 |
+
"eval_runtime": 306.2849,
|
| 38 |
+
"eval_samples_per_second": 12.074,
|
| 39 |
+
"eval_steps_per_second": 12.074,
|
| 40 |
+
"eval_wer": 0.5879318172437987,
|
| 41 |
+
"step": 600
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.10248526774276198,
|
| 45 |
+
"grad_norm": 25.825851440429688,
|
| 46 |
+
"learning_rate": 4.942814232902033e-05,
|
| 47 |
+
"loss": 1.0692,
|
| 48 |
+
"step": 900
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.10248526774276198,
|
| 52 |
+
"eval_loss": 0.9110434055328369,
|
| 53 |
+
"eval_runtime": 299.9179,
|
| 54 |
+
"eval_samples_per_second": 12.33,
|
| 55 |
+
"eval_steps_per_second": 12.33,
|
| 56 |
+
"eval_wer": 0.4886293284906517,
|
| 57 |
+
"step": 900
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.13664702365701598,
|
| 61 |
+
"grad_norm": 34.84864044189453,
|
| 62 |
+
"learning_rate": 4.899491682070241e-05,
|
| 63 |
+
"loss": 0.9443,
|
| 64 |
+
"step": 1200
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.13664702365701598,
|
| 68 |
+
"eval_loss": 0.7600523233413696,
|
| 69 |
+
"eval_runtime": 300.4784,
|
| 70 |
+
"eval_samples_per_second": 12.307,
|
| 71 |
+
"eval_steps_per_second": 12.307,
|
| 72 |
+
"eval_wer": 0.4726567336662677,
|
| 73 |
+
"step": 1200
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.17080877957126997,
|
| 77 |
+
"grad_norm": 80.90399932861328,
|
| 78 |
+
"learning_rate": 4.856169131238448e-05,
|
| 79 |
+
"loss": 0.8235,
|
| 80 |
+
"step": 1500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.17080877957126997,
|
| 84 |
+
"eval_loss": 0.7760590314865112,
|
| 85 |
+
"eval_runtime": 301.4362,
|
| 86 |
+
"eval_samples_per_second": 12.268,
|
| 87 |
+
"eval_steps_per_second": 12.268,
|
| 88 |
+
"eval_wer": 0.3972718642948533,
|
| 89 |
+
"step": 1500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.20497053548552396,
|
| 93 |
+
"grad_norm": 41.970706939697266,
|
| 94 |
+
"learning_rate": 4.8128465804066546e-05,
|
| 95 |
+
"loss": 0.8155,
|
| 96 |
+
"step": 1800
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.20497053548552396,
|
| 100 |
+
"eval_loss": 0.7084236741065979,
|
| 101 |
+
"eval_runtime": 300.2268,
|
| 102 |
+
"eval_samples_per_second": 12.317,
|
| 103 |
+
"eval_steps_per_second": 12.317,
|
| 104 |
+
"eval_wer": 0.40218333402121426,
|
| 105 |
+
"step": 1800
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.23913229139977796,
|
| 109 |
+
"grad_norm": 29.83930206298828,
|
| 110 |
+
"learning_rate": 4.769668438077634e-05,
|
| 111 |
+
"loss": 0.767,
|
| 112 |
+
"step": 2100
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.23913229139977796,
|
| 116 |
+
"eval_loss": 0.6251249313354492,
|
| 117 |
+
"eval_runtime": 298.3691,
|
| 118 |
+
"eval_samples_per_second": 12.394,
|
| 119 |
+
"eval_steps_per_second": 12.394,
|
| 120 |
+
"eval_wer": 0.37560361550208426,
|
| 121 |
+
"step": 2100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.27329404731403195,
|
| 125 |
+
"grad_norm": 203.37936401367188,
|
| 126 |
+
"learning_rate": 4.726345887245841e-05,
|
| 127 |
+
"loss": 0.7517,
|
| 128 |
+
"step": 2400
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.27329404731403195,
|
| 132 |
+
"eval_loss": 0.6125071048736572,
|
| 133 |
+
"eval_runtime": 297.9477,
|
| 134 |
+
"eval_samples_per_second": 12.412,
|
| 135 |
+
"eval_steps_per_second": 12.412,
|
| 136 |
+
"eval_wer": 0.36536794750092866,
|
| 137 |
+
"step": 2400
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.3074558032282859,
|
| 141 |
+
"grad_norm": 15.38986873626709,
|
| 142 |
+
"learning_rate": 4.683023336414049e-05,
|
| 143 |
+
"loss": 0.687,
|
| 144 |
+
"step": 2700
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.3074558032282859,
|
| 148 |
+
"eval_loss": 0.5848016142845154,
|
| 149 |
+
"eval_runtime": 299.5504,
|
| 150 |
+
"eval_samples_per_second": 12.345,
|
| 151 |
+
"eval_steps_per_second": 12.345,
|
| 152 |
+
"eval_wer": 0.3439266994097982,
|
| 153 |
+
"step": 2700
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.34161755914253994,
|
| 157 |
+
"grad_norm": 32.728450775146484,
|
| 158 |
+
"learning_rate": 4.6397007855822556e-05,
|
| 159 |
+
"loss": 0.6509,
|
| 160 |
+
"step": 3000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.34161755914253994,
|
| 164 |
+
"eval_loss": 0.564321756362915,
|
| 165 |
+
"eval_runtime": 300.1274,
|
| 166 |
+
"eval_samples_per_second": 12.321,
|
| 167 |
+
"eval_steps_per_second": 12.321,
|
| 168 |
+
"eval_wer": 0.328160468859631,
|
| 169 |
+
"step": 3000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.3757793150567939,
|
| 173 |
+
"grad_norm": 30.201148986816406,
|
| 174 |
+
"learning_rate": 4.5963782347504624e-05,
|
| 175 |
+
"loss": 0.6632,
|
| 176 |
+
"step": 3300
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.3757793150567939,
|
| 180 |
+
"eval_loss": 0.5509063005447388,
|
| 181 |
+
"eval_runtime": 297.3553,
|
| 182 |
+
"eval_samples_per_second": 12.436,
|
| 183 |
+
"eval_steps_per_second": 12.436,
|
| 184 |
+
"eval_wer": 0.3199265343183788,
|
| 185 |
+
"step": 3300
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.4099410709710479,
|
| 189 |
+
"grad_norm": 105.19284057617188,
|
| 190 |
+
"learning_rate": 4.553055683918669e-05,
|
| 191 |
+
"loss": 0.6108,
|
| 192 |
+
"step": 3600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4099410709710479,
|
| 196 |
+
"eval_loss": 0.5392822623252869,
|
| 197 |
+
"eval_runtime": 300.6256,
|
| 198 |
+
"eval_samples_per_second": 12.301,
|
| 199 |
+
"eval_steps_per_second": 12.301,
|
| 200 |
+
"eval_wer": 0.3341243963844979,
|
| 201 |
+
"step": 3600
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.4441028268853019,
|
| 205 |
+
"grad_norm": 4.016674041748047,
|
| 206 |
+
"learning_rate": 4.5097331330868766e-05,
|
| 207 |
+
"loss": 0.5898,
|
| 208 |
+
"step": 3900
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.4441028268853019,
|
| 212 |
+
"eval_loss": 0.5222619771957397,
|
| 213 |
+
"eval_runtime": 298.4266,
|
| 214 |
+
"eval_samples_per_second": 12.392,
|
| 215 |
+
"eval_steps_per_second": 12.392,
|
| 216 |
+
"eval_wer": 0.32772710388377563,
|
| 217 |
+
"step": 3900
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.4782645827995559,
|
| 221 |
+
"grad_norm": 27.7945613861084,
|
| 222 |
+
"learning_rate": 4.4664105822550834e-05,
|
| 223 |
+
"loss": 0.595,
|
| 224 |
+
"step": 4200
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.4782645827995559,
|
| 228 |
+
"eval_loss": 0.5199422836303711,
|
| 229 |
+
"eval_runtime": 301.9933,
|
| 230 |
+
"eval_samples_per_second": 12.245,
|
| 231 |
+
"eval_steps_per_second": 12.245,
|
| 232 |
+
"eval_wer": 0.31996780717322215,
|
| 233 |
+
"step": 4200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5124263387138099,
|
| 237 |
+
"grad_norm": 37.06507873535156,
|
| 238 |
+
"learning_rate": 4.42308803142329e-05,
|
| 239 |
+
"loss": 0.5644,
|
| 240 |
+
"step": 4500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5124263387138099,
|
| 244 |
+
"eval_loss": 0.5507590174674988,
|
| 245 |
+
"eval_runtime": 302.7901,
|
| 246 |
+
"eval_samples_per_second": 12.213,
|
| 247 |
+
"eval_steps_per_second": 12.213,
|
| 248 |
+
"eval_wer": 0.29186099302488755,
|
| 249 |
+
"step": 4500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.5465880946280639,
|
| 253 |
+
"grad_norm": 62.698890686035156,
|
| 254 |
+
"learning_rate": 4.37990988909427e-05,
|
| 255 |
+
"loss": 0.5787,
|
| 256 |
+
"step": 4800
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.5465880946280639,
|
| 260 |
+
"eval_loss": 0.4994393587112427,
|
| 261 |
+
"eval_runtime": 301.2905,
|
| 262 |
+
"eval_samples_per_second": 12.274,
|
| 263 |
+
"eval_steps_per_second": 12.274,
|
| 264 |
+
"eval_wer": 0.3059763093813199,
|
| 265 |
+
"step": 4800
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.5807498505423179,
|
| 269 |
+
"grad_norm": 18.494293212890625,
|
| 270 |
+
"learning_rate": 4.336587338262477e-05,
|
| 271 |
+
"loss": 0.5752,
|
| 272 |
+
"step": 5100
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5807498505423179,
|
| 276 |
+
"eval_loss": 0.49656036496162415,
|
| 277 |
+
"eval_runtime": 303.7733,
|
| 278 |
+
"eval_samples_per_second": 12.174,
|
| 279 |
+
"eval_steps_per_second": 12.174,
|
| 280 |
+
"eval_wer": 0.29966156259028437,
|
| 281 |
+
"step": 5100
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.6149116064565718,
|
| 285 |
+
"grad_norm": 38.18902587890625,
|
| 286 |
+
"learning_rate": 4.2932647874306844e-05,
|
| 287 |
+
"loss": 0.5353,
|
| 288 |
+
"step": 5400
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.6149116064565718,
|
| 292 |
+
"eval_loss": 0.47309231758117676,
|
| 293 |
+
"eval_runtime": 302.8595,
|
| 294 |
+
"eval_samples_per_second": 12.21,
|
| 295 |
+
"eval_steps_per_second": 12.21,
|
| 296 |
+
"eval_wer": 0.3237030005365471,
|
| 297 |
+
"step": 5400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.6490733623708259,
|
| 301 |
+
"grad_norm": 15.917391777038574,
|
| 302 |
+
"learning_rate": 4.249942236598891e-05,
|
| 303 |
+
"loss": 0.5473,
|
| 304 |
+
"step": 5700
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.6490733623708259,
|
| 308 |
+
"eval_loss": 0.46649235486984253,
|
| 309 |
+
"eval_runtime": 304.2045,
|
| 310 |
+
"eval_samples_per_second": 12.156,
|
| 311 |
+
"eval_steps_per_second": 12.156,
|
| 312 |
+
"eval_wer": 0.3062033100829584,
|
| 313 |
+
"step": 5700
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.6832351182850799,
|
| 317 |
+
"grad_norm": 17.125015258789062,
|
| 318 |
+
"learning_rate": 4.206619685767098e-05,
|
| 319 |
+
"loss": 0.5498,
|
| 320 |
+
"step": 6000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.6832351182850799,
|
| 324 |
+
"eval_loss": 0.48897644877433777,
|
| 325 |
+
"eval_runtime": 305.5464,
|
| 326 |
+
"eval_samples_per_second": 12.103,
|
| 327 |
+
"eval_steps_per_second": 12.103,
|
| 328 |
+
"eval_wer": 0.2875892525485988,
|
| 329 |
+
"step": 6000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.7173968741993338,
|
| 333 |
+
"grad_norm": 18.530088424682617,
|
| 334 |
+
"learning_rate": 4.1632971349353054e-05,
|
| 335 |
+
"loss": 0.5146,
|
| 336 |
+
"step": 6300
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.7173968741993338,
|
| 340 |
+
"eval_loss": 0.4747436046600342,
|
| 341 |
+
"eval_runtime": 304.4465,
|
| 342 |
+
"eval_samples_per_second": 12.147,
|
| 343 |
+
"eval_steps_per_second": 12.147,
|
| 344 |
+
"eval_wer": 0.2926451772669115,
|
| 345 |
+
"step": 6300
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7515586301135878,
|
| 349 |
+
"grad_norm": 24.51532554626465,
|
| 350 |
+
"learning_rate": 4.119974584103512e-05,
|
| 351 |
+
"loss": 0.5398,
|
| 352 |
+
"step": 6600
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7515586301135878,
|
| 356 |
+
"eval_loss": 0.45806387066841125,
|
| 357 |
+
"eval_runtime": 303.5016,
|
| 358 |
+
"eval_samples_per_second": 12.184,
|
| 359 |
+
"eval_steps_per_second": 12.184,
|
| 360 |
+
"eval_wer": 0.29072598951669487,
|
| 361 |
+
"step": 6600
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.7857203860278418,
|
| 365 |
+
"grad_norm": 18.871652603149414,
|
| 366 |
+
"learning_rate": 4.076652033271719e-05,
|
| 367 |
+
"loss": 0.5154,
|
| 368 |
+
"step": 6900
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.7857203860278418,
|
| 372 |
+
"eval_loss": 0.4556685984134674,
|
| 373 |
+
"eval_runtime": 301.5962,
|
| 374 |
+
"eval_samples_per_second": 12.261,
|
| 375 |
+
"eval_steps_per_second": 12.261,
|
| 376 |
+
"eval_wer": 0.29951710759833255,
|
| 377 |
+
"step": 6900
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.8198821419420959,
|
| 381 |
+
"grad_norm": 80.3409194946289,
|
| 382 |
+
"learning_rate": 4.0333294824399264e-05,
|
| 383 |
+
"loss": 0.5386,
|
| 384 |
+
"step": 7200
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.8198821419420959,
|
| 388 |
+
"eval_loss": 0.45153045654296875,
|
| 389 |
+
"eval_runtime": 302.4297,
|
| 390 |
+
"eval_samples_per_second": 12.228,
|
| 391 |
+
"eval_steps_per_second": 12.228,
|
| 392 |
+
"eval_wer": 0.2947500928639234,
|
| 393 |
+
"step": 7200
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.8540438978563498,
|
| 397 |
+
"grad_norm": 4.5673346519470215,
|
| 398 |
+
"learning_rate": 3.990006931608133e-05,
|
| 399 |
+
"loss": 0.5037,
|
| 400 |
+
"step": 7500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.8540438978563498,
|
| 404 |
+
"eval_loss": 0.44558125734329224,
|
| 405 |
+
"eval_runtime": 306.0244,
|
| 406 |
+
"eval_samples_per_second": 12.084,
|
| 407 |
+
"eval_steps_per_second": 12.084,
|
| 408 |
+
"eval_wer": 0.2961120970737546,
|
| 409 |
+
"step": 7500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.8882056537706038,
|
| 413 |
+
"grad_norm": 11.731285095214844,
|
| 414 |
+
"learning_rate": 3.9466843807763406e-05,
|
| 415 |
+
"loss": 0.5344,
|
| 416 |
+
"step": 7800
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.8882056537706038,
|
| 420 |
+
"eval_loss": 0.45088666677474976,
|
| 421 |
+
"eval_runtime": 303.7688,
|
| 422 |
+
"eval_samples_per_second": 12.174,
|
| 423 |
+
"eval_steps_per_second": 12.174,
|
| 424 |
+
"eval_wer": 0.2987741962111519,
|
| 425 |
+
"step": 7800
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.9223674096848578,
|
| 429 |
+
"grad_norm": 21.575855255126953,
|
| 430 |
+
"learning_rate": 3.90350623844732e-05,
|
| 431 |
+
"loss": 0.501,
|
| 432 |
+
"step": 8100
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.9223674096848578,
|
| 436 |
+
"eval_loss": 0.44363677501678467,
|
| 437 |
+
"eval_runtime": 305.2285,
|
| 438 |
+
"eval_samples_per_second": 12.116,
|
| 439 |
+
"eval_steps_per_second": 12.116,
|
| 440 |
+
"eval_wer": 0.2710594741838293,
|
| 441 |
+
"step": 8100
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.9565291655991118,
|
| 445 |
+
"grad_norm": 89.06358337402344,
|
| 446 |
+
"learning_rate": 3.860183687615527e-05,
|
| 447 |
+
"loss": 0.487,
|
| 448 |
+
"step": 8400
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.9565291655991118,
|
| 452 |
+
"eval_loss": 0.42328470945358276,
|
| 453 |
+
"eval_runtime": 303.3048,
|
| 454 |
+
"eval_samples_per_second": 12.192,
|
| 455 |
+
"eval_steps_per_second": 12.192,
|
| 456 |
+
"eval_wer": 0.27491848611168435,
|
| 457 |
+
"step": 8400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9906909215133658,
|
| 461 |
+
"grad_norm": 2.7736120223999023,
|
| 462 |
+
"learning_rate": 3.816861136783734e-05,
|
| 463 |
+
"loss": 0.4692,
|
| 464 |
+
"step": 8700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.9906909215133658,
|
| 468 |
+
"eval_loss": 0.46611106395721436,
|
| 469 |
+
"eval_runtime": 303.8834,
|
| 470 |
+
"eval_samples_per_second": 12.169,
|
| 471 |
+
"eval_steps_per_second": 12.169,
|
| 472 |
+
"eval_wer": 0.25322960089149366,
|
| 473 |
+
"step": 8700
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.0248526774276199,
|
| 477 |
+
"grad_norm": 2.908313035964966,
|
| 478 |
+
"learning_rate": 3.773538585951941e-05,
|
| 479 |
+
"loss": 0.462,
|
| 480 |
+
"step": 9000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.0248526774276199,
|
| 484 |
+
"eval_loss": 0.4197135269641876,
|
| 485 |
+
"eval_runtime": 302.465,
|
| 486 |
+
"eval_samples_per_second": 12.226,
|
| 487 |
+
"eval_steps_per_second": 12.226,
|
| 488 |
+
"eval_wer": 0.2722770234017087,
|
| 489 |
+
"step": 9000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.0590144333418738,
|
| 493 |
+
"grad_norm": 1.9524333477020264,
|
| 494 |
+
"learning_rate": 3.7302160351201484e-05,
|
| 495 |
+
"loss": 0.4508,
|
| 496 |
+
"step": 9300
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.0590144333418738,
|
| 500 |
+
"eval_loss": 0.4316381514072418,
|
| 501 |
+
"eval_runtime": 304.5385,
|
| 502 |
+
"eval_samples_per_second": 12.143,
|
| 503 |
+
"eval_steps_per_second": 12.143,
|
| 504 |
+
"eval_wer": 0.25838870774691486,
|
| 505 |
+
"step": 9300
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.0931761892561278,
|
| 509 |
+
"grad_norm": 2.584221124649048,
|
| 510 |
+
"learning_rate": 3.686893484288355e-05,
|
| 511 |
+
"loss": 0.4702,
|
| 512 |
+
"step": 9600
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0931761892561278,
|
| 516 |
+
"eval_loss": 0.4147655963897705,
|
| 517 |
+
"eval_runtime": 302.8309,
|
| 518 |
+
"eval_samples_per_second": 12.211,
|
| 519 |
+
"eval_steps_per_second": 12.211,
|
| 520 |
+
"eval_wer": 0.2689132857319741,
|
| 521 |
+
"step": 9600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.1273379451703818,
|
| 525 |
+
"grad_norm": 1.6557509899139404,
|
| 526 |
+
"learning_rate": 3.643570933456562e-05,
|
| 527 |
+
"loss": 0.4517,
|
| 528 |
+
"step": 9900
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.1273379451703818,
|
| 532 |
+
"eval_loss": 0.39497703313827515,
|
| 533 |
+
"eval_runtime": 304.7729,
|
| 534 |
+
"eval_samples_per_second": 12.134,
|
| 535 |
+
"eval_steps_per_second": 12.134,
|
| 536 |
+
"eval_wer": 0.2549217879400718,
|
| 537 |
+
"step": 9900
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.1614997010846357,
|
| 541 |
+
"grad_norm": 1.9434764385223389,
|
| 542 |
+
"learning_rate": 3.6002483826247694e-05,
|
| 543 |
+
"loss": 0.4408,
|
| 544 |
+
"step": 10200
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.1614997010846357,
|
| 548 |
+
"eval_loss": 0.43082132935523987,
|
| 549 |
+
"eval_runtime": 303.9383,
|
| 550 |
+
"eval_samples_per_second": 12.167,
|
| 551 |
+
"eval_steps_per_second": 12.167,
|
| 552 |
+
"eval_wer": 0.25514878864171037,
|
| 553 |
+
"step": 10200
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.1956614569988897,
|
| 557 |
+
"grad_norm": 0.8506720066070557,
|
| 558 |
+
"learning_rate": 3.556925831792976e-05,
|
| 559 |
+
"loss": 0.4636,
|
| 560 |
+
"step": 10500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.1956614569988897,
|
| 564 |
+
"eval_loss": 0.4033319354057312,
|
| 565 |
+
"eval_runtime": 302.9721,
|
| 566 |
+
"eval_samples_per_second": 12.206,
|
| 567 |
+
"eval_steps_per_second": 12.206,
|
| 568 |
+
"eval_wer": 0.26996574353048003,
|
| 569 |
+
"step": 10500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.2298232129131437,
|
| 573 |
+
"grad_norm": 4.787822246551514,
|
| 574 |
+
"learning_rate": 3.513747689463956e-05,
|
| 575 |
+
"loss": 0.4583,
|
| 576 |
+
"step": 10800
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.2298232129131437,
|
| 580 |
+
"eval_loss": 0.40962740778923035,
|
| 581 |
+
"eval_runtime": 301.5186,
|
| 582 |
+
"eval_samples_per_second": 12.265,
|
| 583 |
+
"eval_steps_per_second": 12.265,
|
| 584 |
+
"eval_wer": 0.2555821536175657,
|
| 585 |
+
"step": 10800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.2639849688273976,
|
| 589 |
+
"grad_norm": 6.576323986053467,
|
| 590 |
+
"learning_rate": 3.470425138632163e-05,
|
| 591 |
+
"loss": 0.4315,
|
| 592 |
+
"step": 11100
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.2639849688273976,
|
| 596 |
+
"eval_loss": 0.3883008658885956,
|
| 597 |
+
"eval_runtime": 299.5319,
|
| 598 |
+
"eval_samples_per_second": 12.346,
|
| 599 |
+
"eval_steps_per_second": 12.346,
|
| 600 |
+
"eval_wer": 0.26812910148995006,
|
| 601 |
+
"step": 11100
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.2981467247416516,
|
| 605 |
+
"grad_norm": 1.379813313484192,
|
| 606 |
+
"learning_rate": 3.427246996303142e-05,
|
| 607 |
+
"loss": 0.4172,
|
| 608 |
+
"step": 11400
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.2981467247416516,
|
| 612 |
+
"eval_loss": 0.37370720505714417,
|
| 613 |
+
"eval_runtime": 301.6675,
|
| 614 |
+
"eval_samples_per_second": 12.259,
|
| 615 |
+
"eval_steps_per_second": 12.259,
|
| 616 |
+
"eval_wer": 0.2529200544801684,
|
| 617 |
+
"step": 11400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.3323084806559058,
|
| 621 |
+
"grad_norm": 2.169112205505371,
|
| 622 |
+
"learning_rate": 3.3839244454713497e-05,
|
| 623 |
+
"loss": 0.4177,
|
| 624 |
+
"step": 11700
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.3323084806559058,
|
| 628 |
+
"eval_loss": 0.399240642786026,
|
| 629 |
+
"eval_runtime": 301.556,
|
| 630 |
+
"eval_samples_per_second": 12.263,
|
| 631 |
+
"eval_steps_per_second": 12.263,
|
| 632 |
+
"eval_wer": 0.24724503693920508,
|
| 633 |
+
"step": 11700
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.3664702365701598,
|
| 637 |
+
"grad_norm": 1.855020523071289,
|
| 638 |
+
"learning_rate": 3.3406018946395564e-05,
|
| 639 |
+
"loss": 0.3975,
|
| 640 |
+
"step": 12000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.3664702365701598,
|
| 644 |
+
"eval_loss": 0.3716418743133545,
|
| 645 |
+
"eval_runtime": 301.4037,
|
| 646 |
+
"eval_samples_per_second": 12.269,
|
| 647 |
+
"eval_steps_per_second": 12.269,
|
| 648 |
+
"eval_wer": 0.24852449543934954,
|
| 649 |
+
"step": 12000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.4006319924844137,
|
| 653 |
+
"grad_norm": 3.5865867137908936,
|
| 654 |
+
"learning_rate": 3.297279343807764e-05,
|
| 655 |
+
"loss": 0.4044,
|
| 656 |
+
"step": 12300
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.4006319924844137,
|
| 660 |
+
"eval_loss": 0.3852696120738983,
|
| 661 |
+
"eval_runtime": 299.7735,
|
| 662 |
+
"eval_samples_per_second": 12.336,
|
| 663 |
+
"eval_steps_per_second": 12.336,
|
| 664 |
+
"eval_wer": 0.25232159808493954,
|
| 665 |
+
"step": 12300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 1.4347937483986677,
|
| 669 |
+
"grad_norm": 7.861583232879639,
|
| 670 |
+
"learning_rate": 3.2539567929759707e-05,
|
| 671 |
+
"loss": 0.4497,
|
| 672 |
+
"step": 12600
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.4347937483986677,
|
| 676 |
+
"eval_loss": 0.37975797057151794,
|
| 677 |
+
"eval_runtime": 303.0367,
|
| 678 |
+
"eval_samples_per_second": 12.203,
|
| 679 |
+
"eval_steps_per_second": 12.203,
|
| 680 |
+
"eval_wer": 0.24646085269718107,
|
| 681 |
+
"step": 12600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.4689555043129217,
|
| 685 |
+
"grad_norm": 2.6698834896087646,
|
| 686 |
+
"learning_rate": 3.2106342421441774e-05,
|
| 687 |
+
"loss": 0.4188,
|
| 688 |
+
"step": 12900
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.4689555043129217,
|
| 692 |
+
"eval_loss": 0.38219255208969116,
|
| 693 |
+
"eval_runtime": 301.4473,
|
| 694 |
+
"eval_samples_per_second": 12.267,
|
| 695 |
+
"eval_steps_per_second": 12.267,
|
| 696 |
+
"eval_wer": 0.24937058896363862,
|
| 697 |
+
"step": 12900
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.5031172602271756,
|
| 701 |
+
"grad_norm": 2.8027360439300537,
|
| 702 |
+
"learning_rate": 3.167311691312385e-05,
|
| 703 |
+
"loss": 0.4424,
|
| 704 |
+
"step": 13200
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.5031172602271756,
|
| 708 |
+
"eval_loss": 0.3559640944004059,
|
| 709 |
+
"eval_runtime": 300.0982,
|
| 710 |
+
"eval_samples_per_second": 12.323,
|
| 711 |
+
"eval_steps_per_second": 12.323,
|
| 712 |
+
"eval_wer": 0.24485121135828966,
|
| 713 |
+
"step": 13200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.5372790161414298,
|
| 717 |
+
"grad_norm": 3.83882212638855,
|
| 718 |
+
"learning_rate": 3.1239891404805917e-05,
|
| 719 |
+
"loss": 0.4249,
|
| 720 |
+
"step": 13500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.5372790161414298,
|
| 724 |
+
"eval_loss": 0.3629530966281891,
|
| 725 |
+
"eval_runtime": 303.9778,
|
| 726 |
+
"eval_samples_per_second": 12.165,
|
| 727 |
+
"eval_steps_per_second": 12.165,
|
| 728 |
+
"eval_wer": 0.25139295885096374,
|
| 729 |
+
"step": 13500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.5714407720556838,
|
| 733 |
+
"grad_norm": 3.2446749210357666,
|
| 734 |
+
"learning_rate": 3.0806665896487984e-05,
|
| 735 |
+
"loss": 0.4287,
|
| 736 |
+
"step": 13800
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.5714407720556838,
|
| 740 |
+
"eval_loss": 0.3662088215351105,
|
| 741 |
+
"eval_runtime": 303.0939,
|
| 742 |
+
"eval_samples_per_second": 12.201,
|
| 743 |
+
"eval_steps_per_second": 12.201,
|
| 744 |
+
"eval_wer": 0.24173511081761526,
|
| 745 |
+
"step": 13800
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.6056025279699377,
|
| 749 |
+
"grad_norm": 1.8408727645874023,
|
| 750 |
+
"learning_rate": 3.0373440388170055e-05,
|
| 751 |
+
"loss": 0.3712,
|
| 752 |
+
"step": 14100
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.6056025279699377,
|
| 756 |
+
"eval_loss": 0.3713747262954712,
|
| 757 |
+
"eval_runtime": 301.435,
|
| 758 |
+
"eval_samples_per_second": 12.268,
|
| 759 |
+
"eval_steps_per_second": 12.268,
|
| 760 |
+
"eval_wer": 0.2561806100127946,
|
| 761 |
+
"step": 14100
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.6397642838841917,
|
| 765 |
+
"grad_norm": 1.50343656539917,
|
| 766 |
+
"learning_rate": 2.994021487985213e-05,
|
| 767 |
+
"loss": 0.3893,
|
| 768 |
+
"step": 14400
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.6397642838841917,
|
| 772 |
+
"eval_loss": 0.3711312413215637,
|
| 773 |
+
"eval_runtime": 300.5222,
|
| 774 |
+
"eval_samples_per_second": 12.305,
|
| 775 |
+
"eval_steps_per_second": 12.305,
|
| 776 |
+
"eval_wer": 0.23325353914730282,
|
| 777 |
+
"step": 14400
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 1.6739260397984457,
|
| 781 |
+
"grad_norm": 2.6718335151672363,
|
| 782 |
+
"learning_rate": 2.95069893715342e-05,
|
| 783 |
+
"loss": 0.3935,
|
| 784 |
+
"step": 14700
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 1.6739260397984457,
|
| 788 |
+
"eval_loss": 0.3715294897556305,
|
| 789 |
+
"eval_runtime": 301.8378,
|
| 790 |
+
"eval_samples_per_second": 12.252,
|
| 791 |
+
"eval_steps_per_second": 12.252,
|
| 792 |
+
"eval_wer": 0.2412604729869165,
|
| 793 |
+
"step": 14700
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.7080877957126996,
|
| 797 |
+
"grad_norm": 1.3738945722579956,
|
| 798 |
+
"learning_rate": 2.907376386321627e-05,
|
| 799 |
+
"loss": 0.3982,
|
| 800 |
+
"step": 15000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.7080877957126996,
|
| 804 |
+
"eval_loss": 0.3551383316516876,
|
| 805 |
+
"eval_runtime": 302.7384,
|
| 806 |
+
"eval_samples_per_second": 12.215,
|
| 807 |
+
"eval_steps_per_second": 12.215,
|
| 808 |
+
"eval_wer": 0.24821494902802427,
|
| 809 |
+
"step": 15000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.7422495516269536,
|
| 813 |
+
"grad_norm": 2.5263662338256836,
|
| 814 |
+
"learning_rate": 2.864053835489834e-05,
|
| 815 |
+
"loss": 0.4124,
|
| 816 |
+
"step": 15300
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.7422495516269536,
|
| 820 |
+
"eval_loss": 0.3519354462623596,
|
| 821 |
+
"eval_runtime": 303.939,
|
| 822 |
+
"eval_samples_per_second": 12.167,
|
| 823 |
+
"eval_steps_per_second": 12.167,
|
| 824 |
+
"eval_wer": 0.24121920013207312,
|
| 825 |
+
"step": 15300
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.7764113075412076,
|
| 829 |
+
"grad_norm": 1.8874987363815308,
|
| 830 |
+
"learning_rate": 2.8207312846580408e-05,
|
| 831 |
+
"loss": 0.3853,
|
| 832 |
+
"step": 15600
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.7764113075412076,
|
| 836 |
+
"eval_loss": 0.3428538739681244,
|
| 837 |
+
"eval_runtime": 303.4637,
|
| 838 |
+
"eval_samples_per_second": 12.186,
|
| 839 |
+
"eval_steps_per_second": 12.186,
|
| 840 |
+
"eval_wer": 0.2417970200998803,
|
| 841 |
+
"step": 15600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.8105730634554615,
|
| 845 |
+
"grad_norm": 1.8015034198760986,
|
| 846 |
+
"learning_rate": 2.777408733826248e-05,
|
| 847 |
+
"loss": 0.4096,
|
| 848 |
+
"step": 15900
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.8105730634554615,
|
| 852 |
+
"eval_loss": 0.34074389934539795,
|
| 853 |
+
"eval_runtime": 301.0523,
|
| 854 |
+
"eval_samples_per_second": 12.284,
|
| 855 |
+
"eval_steps_per_second": 12.284,
|
| 856 |
+
"eval_wer": 0.23938255809154318,
|
| 857 |
+
"step": 15900
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.8447348193697155,
|
| 861 |
+
"grad_norm": 1.9667861461639404,
|
| 862 |
+
"learning_rate": 2.7340861829944547e-05,
|
| 863 |
+
"loss": 0.3816,
|
| 864 |
+
"step": 16200
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.8447348193697155,
|
| 868 |
+
"eval_loss": 0.360734224319458,
|
| 869 |
+
"eval_runtime": 301.5423,
|
| 870 |
+
"eval_samples_per_second": 12.264,
|
| 871 |
+
"eval_steps_per_second": 12.264,
|
| 872 |
+
"eval_wer": 0.23700936893804944,
|
| 873 |
+
"step": 16200
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 1.8788965752839695,
|
| 877 |
+
"grad_norm": 1.8239201307296753,
|
| 878 |
+
"learning_rate": 2.6909080406654346e-05,
|
| 879 |
+
"loss": 0.3769,
|
| 880 |
+
"step": 16500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 1.8788965752839695,
|
| 884 |
+
"eval_loss": 0.36014851927757263,
|
| 885 |
+
"eval_runtime": 301.5685,
|
| 886 |
+
"eval_samples_per_second": 12.263,
|
| 887 |
+
"eval_steps_per_second": 12.263,
|
| 888 |
+
"eval_wer": 0.22910561723554418,
|
| 889 |
+
"step": 16500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 1.9130583311982234,
|
| 893 |
+
"grad_norm": 1.5791462659835815,
|
| 894 |
+
"learning_rate": 2.6475854898336417e-05,
|
| 895 |
+
"loss": 0.3428,
|
| 896 |
+
"step": 16800
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.9130583311982234,
|
| 900 |
+
"eval_loss": 0.3577890694141388,
|
| 901 |
+
"eval_runtime": 299.2343,
|
| 902 |
+
"eval_samples_per_second": 12.358,
|
| 903 |
+
"eval_steps_per_second": 12.358,
|
| 904 |
+
"eval_wer": 0.22830079656609847,
|
| 905 |
+
"step": 16800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.9472200871124776,
|
| 909 |
+
"grad_norm": 2.0596249103546143,
|
| 910 |
+
"learning_rate": 2.6042629390018485e-05,
|
| 911 |
+
"loss": 0.3636,
|
| 912 |
+
"step": 17100
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.9472200871124776,
|
| 916 |
+
"eval_loss": 0.348527193069458,
|
| 917 |
+
"eval_runtime": 301.456,
|
| 918 |
+
"eval_samples_per_second": 12.267,
|
| 919 |
+
"eval_steps_per_second": 12.267,
|
| 920 |
+
"eval_wer": 0.2333979941392546,
|
| 921 |
+
"step": 17100
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 1.9813818430267316,
|
| 925 |
+
"grad_norm": 1.8134913444519043,
|
| 926 |
+
"learning_rate": 2.5609403881700556e-05,
|
| 927 |
+
"loss": 0.3594,
|
| 928 |
+
"step": 17400
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.9813818430267316,
|
| 932 |
+
"eval_loss": 0.3539000451564789,
|
| 933 |
+
"eval_runtime": 299.8125,
|
| 934 |
+
"eval_samples_per_second": 12.334,
|
| 935 |
+
"eval_steps_per_second": 12.334,
|
| 936 |
+
"eval_wer": 0.23412026909901357,
|
| 937 |
+
"step": 17400
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.015543598940986,
|
| 941 |
+
"grad_norm": 4.158298969268799,
|
| 942 |
+
"learning_rate": 2.5176178373382624e-05,
|
| 943 |
+
"loss": 0.3692,
|
| 944 |
+
"step": 17700
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.015543598940986,
|
| 948 |
+
"eval_loss": 0.33834952116012573,
|
| 949 |
+
"eval_runtime": 300.4735,
|
| 950 |
+
"eval_samples_per_second": 12.307,
|
| 951 |
+
"eval_steps_per_second": 12.307,
|
| 952 |
+
"eval_wer": 0.22817697800156836,
|
| 953 |
+
"step": 17700
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 2.0497053548552397,
|
| 957 |
+
"grad_norm": 1.628749132156372,
|
| 958 |
+
"learning_rate": 2.47429528650647e-05,
|
| 959 |
+
"loss": 0.3295,
|
| 960 |
+
"step": 18000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 2.0497053548552397,
|
| 964 |
+
"eval_loss": 0.33542123436927795,
|
| 965 |
+
"eval_runtime": 299.4927,
|
| 966 |
+
"eval_samples_per_second": 12.348,
|
| 967 |
+
"eval_steps_per_second": 12.348,
|
| 968 |
+
"eval_wer": 0.23740146105906146,
|
| 969 |
+
"step": 18000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.0838671107694937,
|
| 973 |
+
"grad_norm": 1.8399901390075684,
|
| 974 |
+
"learning_rate": 2.4309727356746766e-05,
|
| 975 |
+
"loss": 0.3442,
|
| 976 |
+
"step": 18300
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.0838671107694937,
|
| 980 |
+
"eval_loss": 0.33928370475769043,
|
| 981 |
+
"eval_runtime": 299.4918,
|
| 982 |
+
"eval_samples_per_second": 12.348,
|
| 983 |
+
"eval_steps_per_second": 12.348,
|
| 984 |
+
"eval_wer": 0.23403772338932685,
|
| 985 |
+
"step": 18300
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.1180288666837477,
|
| 989 |
+
"grad_norm": 2.220979928970337,
|
| 990 |
+
"learning_rate": 2.3876501848428837e-05,
|
| 991 |
+
"loss": 0.3306,
|
| 992 |
+
"step": 18600
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.1180288666837477,
|
| 996 |
+
"eval_loss": 0.35673782229423523,
|
| 997 |
+
"eval_runtime": 303.0121,
|
| 998 |
+
"eval_samples_per_second": 12.204,
|
| 999 |
+
"eval_steps_per_second": 12.204,
|
| 1000 |
+
"eval_wer": 0.2381650088736638,
|
| 1001 |
+
"step": 18600
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.1521906225980016,
|
| 1005 |
+
"grad_norm": 2.257617235183716,
|
| 1006 |
+
"learning_rate": 2.3443276340110905e-05,
|
| 1007 |
+
"loss": 0.3243,
|
| 1008 |
+
"step": 18900
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 2.1521906225980016,
|
| 1012 |
+
"eval_loss": 0.3409930169582367,
|
| 1013 |
+
"eval_runtime": 299.9306,
|
| 1014 |
+
"eval_samples_per_second": 12.33,
|
| 1015 |
+
"eval_steps_per_second": 12.33,
|
| 1016 |
+
"eval_wer": 0.2286722522596888,
|
| 1017 |
+
"step": 18900
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.1863523785122556,
|
| 1021 |
+
"grad_norm": 4.057617664337158,
|
| 1022 |
+
"learning_rate": 2.3010050831792976e-05,
|
| 1023 |
+
"loss": 0.3426,
|
| 1024 |
+
"step": 19200
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.1863523785122556,
|
| 1028 |
+
"eval_loss": 0.3244094252586365,
|
| 1029 |
+
"eval_runtime": 302.9284,
|
| 1030 |
+
"eval_samples_per_second": 12.208,
|
| 1031 |
+
"eval_steps_per_second": 12.208,
|
| 1032 |
+
"eval_wer": 0.23230426348590533,
|
| 1033 |
+
"step": 19200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.2205141344265096,
|
| 1037 |
+
"grad_norm": 4.836043357849121,
|
| 1038 |
+
"learning_rate": 2.2576825323475047e-05,
|
| 1039 |
+
"loss": 0.3552,
|
| 1040 |
+
"step": 19500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.2205141344265096,
|
| 1044 |
+
"eval_loss": 0.33564862608909607,
|
| 1045 |
+
"eval_runtime": 300.1908,
|
| 1046 |
+
"eval_samples_per_second": 12.319,
|
| 1047 |
+
"eval_steps_per_second": 12.319,
|
| 1048 |
+
"eval_wer": 0.23176771637294152,
|
| 1049 |
+
"step": 19500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.2546758903407635,
|
| 1053 |
+
"grad_norm": 3.1353209018707275,
|
| 1054 |
+
"learning_rate": 2.214359981515712e-05,
|
| 1055 |
+
"loss": 0.3558,
|
| 1056 |
+
"step": 19800
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 2.2546758903407635,
|
| 1060 |
+
"eval_loss": 0.3686023950576782,
|
| 1061 |
+
"eval_runtime": 300.6653,
|
| 1062 |
+
"eval_samples_per_second": 12.299,
|
| 1063 |
+
"eval_steps_per_second": 12.299,
|
| 1064 |
+
"eval_wer": 0.22246068760576168,
|
| 1065 |
+
"step": 19800
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.2888376462550175,
|
| 1069 |
+
"grad_norm": 1.1471089124679565,
|
| 1070 |
+
"learning_rate": 2.1710374306839186e-05,
|
| 1071 |
+
"loss": 0.3485,
|
| 1072 |
+
"step": 20100
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.2888376462550175,
|
| 1076 |
+
"eval_loss": 0.34854745864868164,
|
| 1077 |
+
"eval_runtime": 301.8158,
|
| 1078 |
+
"eval_samples_per_second": 12.253,
|
| 1079 |
+
"eval_steps_per_second": 12.253,
|
| 1080 |
+
"eval_wer": 0.2229765982913038,
|
| 1081 |
+
"step": 20100
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.3229994021692715,
|
| 1085 |
+
"grad_norm": 4.745022773742676,
|
| 1086 |
+
"learning_rate": 2.1277148798521257e-05,
|
| 1087 |
+
"loss": 0.3195,
|
| 1088 |
+
"step": 20400
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.3229994021692715,
|
| 1092 |
+
"eval_loss": 0.31967541575431824,
|
| 1093 |
+
"eval_runtime": 303.1201,
|
| 1094 |
+
"eval_samples_per_second": 12.2,
|
| 1095 |
+
"eval_steps_per_second": 12.2,
|
| 1096 |
+
"eval_wer": 0.2229765982913038,
|
| 1097 |
+
"step": 20400
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.3571611580835254,
|
| 1101 |
+
"grad_norm": 0.8967244625091553,
|
| 1102 |
+
"learning_rate": 2.084392329020333e-05,
|
| 1103 |
+
"loss": 0.3145,
|
| 1104 |
+
"step": 20700
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.3571611580835254,
|
| 1108 |
+
"eval_loss": 0.33116424083709717,
|
| 1109 |
+
"eval_runtime": 298.6697,
|
| 1110 |
+
"eval_samples_per_second": 12.382,
|
| 1111 |
+
"eval_steps_per_second": 12.382,
|
| 1112 |
+
"eval_wer": 0.22941516364686945,
|
| 1113 |
+
"step": 20700
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 2.3913229139977794,
|
| 1117 |
+
"grad_norm": 1.37799870967865,
|
| 1118 |
+
"learning_rate": 2.04106977818854e-05,
|
| 1119 |
+
"loss": 0.3238,
|
| 1120 |
+
"step": 21000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 2.3913229139977794,
|
| 1124 |
+
"eval_loss": 0.3331291675567627,
|
| 1125 |
+
"eval_runtime": 299.3013,
|
| 1126 |
+
"eval_samples_per_second": 12.355,
|
| 1127 |
+
"eval_steps_per_second": 12.355,
|
| 1128 |
+
"eval_wer": 0.22097486483140039,
|
| 1129 |
+
"step": 21000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.4254846699120334,
|
| 1133 |
+
"grad_norm": 3.7956511974334717,
|
| 1134 |
+
"learning_rate": 1.9978916358595196e-05,
|
| 1135 |
+
"loss": 0.3288,
|
| 1136 |
+
"step": 21300
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.4254846699120334,
|
| 1140 |
+
"eval_loss": 0.31718209385871887,
|
| 1141 |
+
"eval_runtime": 302.9796,
|
| 1142 |
+
"eval_samples_per_second": 12.205,
|
| 1143 |
+
"eval_steps_per_second": 12.205,
|
| 1144 |
+
"eval_wer": 0.2272070659127492,
|
| 1145 |
+
"step": 21300
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.4596464258262873,
|
| 1149 |
+
"grad_norm": 2.695967674255371,
|
| 1150 |
+
"learning_rate": 1.9545690850277264e-05,
|
| 1151 |
+
"loss": 0.3398,
|
| 1152 |
+
"step": 21600
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.4596464258262873,
|
| 1156 |
+
"eval_loss": 0.3227500915527344,
|
| 1157 |
+
"eval_runtime": 303.1823,
|
| 1158 |
+
"eval_samples_per_second": 12.197,
|
| 1159 |
+
"eval_steps_per_second": 12.197,
|
| 1160 |
+
"eval_wer": 0.21818894712947295,
|
| 1161 |
+
"step": 21600
|
| 1162 |
+
}
|
| 1163 |
+
],
|
| 1164 |
+
"logging_steps": 300,
|
| 1165 |
+
"max_steps": 35124,
|
| 1166 |
+
"num_input_tokens_seen": 0,
|
| 1167 |
+
"num_train_epochs": 4,
|
| 1168 |
+
"save_steps": 600,
|
| 1169 |
+
"stateful_callbacks": {
|
| 1170 |
+
"TrainerControl": {
|
| 1171 |
+
"args": {
|
| 1172 |
+
"should_epoch_stop": false,
|
| 1173 |
+
"should_evaluate": false,
|
| 1174 |
+
"should_log": false,
|
| 1175 |
+
"should_save": true,
|
| 1176 |
+
"should_training_stop": false
|
| 1177 |
+
},
|
| 1178 |
+
"attributes": {}
|
| 1179 |
+
}
|
| 1180 |
+
},
|
| 1181 |
+
"total_flos": 2.982387127199302e+19,
|
| 1182 |
+
"train_batch_size": 2,
|
| 1183 |
+
"trial_name": null,
|
| 1184 |
+
"trial_params": null
|
| 1185 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5e185addb91073320ac8f20c1b4ffa7d558a32cc6b30c3252573ea642967c89
|
| 3 |
+
size 5048
|