Training in progress, step 1800, checkpoint
Browse files- checkpoint-1800/config.json +82 -0
- checkpoint-1800/model.safetensors +3 -0
- checkpoint-1800/optimizer.pt +3 -0
- checkpoint-1800/preprocessor_config.json +11 -0
- checkpoint-1800/rng_state.pth +3 -0
- checkpoint-1800/scheduler.pt +3 -0
- checkpoint-1800/trainer_state.json +1320 -0
- checkpoint-1800/training_args.bin +3 -0
checkpoint-1800/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": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2BertForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 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": 0.0,
|
| 24 |
+
"feat_quantizer_dropout": 0.0,
|
| 25 |
+
"feature_projection_input_dim": 160,
|
| 26 |
+
"final_dropout": 0.1,
|
| 27 |
+
"hidden_act": "swish",
|
| 28 |
+
"hidden_dropout": 0.0,
|
| 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": 147,
|
| 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": "4.41.2",
|
| 78 |
+
"use_intermediate_ffn_before_adapter": true,
|
| 79 |
+
"use_weighted_layer_sum": false,
|
| 80 |
+
"vocab_size": 150,
|
| 81 |
+
"xvector_output_dim": 512
|
| 82 |
+
}
|
checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e4723e4da159a3c5e1c05164271f929e85dfe822b925e21a1c0a4a130bc5739c
|
| 3 |
+
size 2457004984
|
checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e40e97e8106afc875f3ef370057b954b95cbc54e0da372a1499dc489c688eb4f
|
| 3 |
+
size 4914473078
|
checkpoint-1800/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": 1,
|
| 7 |
+
"processor_class": "Wav2Vec2BertProcessor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000,
|
| 10 |
+
"stride": 2
|
| 11 |
+
}
|
checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49c86962bb776f59a78f2096bb2776e38579a9e3782c352b6b1deb07d1128841
|
| 3 |
+
size 14244
|
checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f144bc56e45be0868a5b63a031364852c6f42d342852625e0cff4fc5b0dc0af0
|
| 3 |
+
size 1064
|
checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1320 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.34055221740156477,
|
| 3 |
+
"best_model_checkpoint": "./checkpoints/w2v-pa-v2/checkpoint-1800",
|
| 4 |
+
"epoch": 0.06541173050367033,
|
| 5 |
+
"eval_steps": 600,
|
| 6 |
+
"global_step": 1800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00036339850279816846,
|
| 13 |
+
"grad_norm": 12.712770462036133,
|
| 14 |
+
"learning_rate": 3.633985027981685e-08,
|
| 15 |
+
"loss": 8.1015,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0007267970055963369,
|
| 20 |
+
"grad_norm": 14.929631233215332,
|
| 21 |
+
"learning_rate": 7.26797005596337e-08,
|
| 22 |
+
"loss": 8.5111,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0010901955083945055,
|
| 27 |
+
"grad_norm": 12.774781227111816,
|
| 28 |
+
"learning_rate": 1.0901955083945056e-07,
|
| 29 |
+
"loss": 8.71,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0014535940111926739,
|
| 34 |
+
"grad_norm": 15.27083969116211,
|
| 35 |
+
"learning_rate": 1.417254160912857e-07,
|
| 36 |
+
"loss": 9.2894,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0018169925139908424,
|
| 41 |
+
"grad_norm": 31.209775924682617,
|
| 42 |
+
"learning_rate": 1.7806526637110256e-07,
|
| 43 |
+
"loss": 8.6779,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.002180391016789011,
|
| 48 |
+
"grad_norm": 12.960335731506348,
|
| 49 |
+
"learning_rate": 2.1440511665091943e-07,
|
| 50 |
+
"loss": 8.0399,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.002543789519587179,
|
| 55 |
+
"grad_norm": 14.452157020568848,
|
| 56 |
+
"learning_rate": 2.5074496693073626e-07,
|
| 57 |
+
"loss": 7.9357,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0029071880223853477,
|
| 62 |
+
"grad_norm": 12.74867057800293,
|
| 63 |
+
"learning_rate": 2.8708481721055307e-07,
|
| 64 |
+
"loss": 8.0764,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0032705865251835163,
|
| 69 |
+
"grad_norm": 19.905397415161133,
|
| 70 |
+
"learning_rate": 3.2342466749036993e-07,
|
| 71 |
+
"loss": 8.2522,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.003633985027981685,
|
| 76 |
+
"grad_norm": Infinity,
|
| 77 |
+
"learning_rate": 3.5613053274220513e-07,
|
| 78 |
+
"loss": 7.8161,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.003997383530779853,
|
| 83 |
+
"grad_norm": 15.877684593200684,
|
| 84 |
+
"learning_rate": 3.92470383022022e-07,
|
| 85 |
+
"loss": 7.4866,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.004360782033578022,
|
| 90 |
+
"grad_norm": 19.216800689697266,
|
| 91 |
+
"learning_rate": 4.2881023330183885e-07,
|
| 92 |
+
"loss": 7.0324,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0047241805363761906,
|
| 97 |
+
"grad_norm": 16.937118530273438,
|
| 98 |
+
"learning_rate": 4.651500835816557e-07,
|
| 99 |
+
"loss": 6.7873,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.005087579039174358,
|
| 104 |
+
"grad_norm": 27.858692169189453,
|
| 105 |
+
"learning_rate": 5.014899338614725e-07,
|
| 106 |
+
"loss": 7.2063,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.005450977541972527,
|
| 111 |
+
"grad_norm": Infinity,
|
| 112 |
+
"learning_rate": 5.341957991133076e-07,
|
| 113 |
+
"loss": 6.5827,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.005814376044770695,
|
| 118 |
+
"grad_norm": 21.252164840698242,
|
| 119 |
+
"learning_rate": 5.669016643651428e-07,
|
| 120 |
+
"loss": 7.819,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.006177774547568864,
|
| 125 |
+
"grad_norm": 20.977886199951172,
|
| 126 |
+
"learning_rate": 6.032415146449597e-07,
|
| 127 |
+
"loss": 5.4741,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0065411730503670325,
|
| 132 |
+
"grad_norm": 35.25390625,
|
| 133 |
+
"learning_rate": 6.395813649247765e-07,
|
| 134 |
+
"loss": 5.7225,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006904571553165201,
|
| 139 |
+
"grad_norm": 7.165033340454102,
|
| 140 |
+
"learning_rate": 6.759212152045934e-07,
|
| 141 |
+
"loss": 4.7475,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.00726797005596337,
|
| 146 |
+
"grad_norm": 14.877301216125488,
|
| 147 |
+
"learning_rate": 7.122610654844103e-07,
|
| 148 |
+
"loss": 4.6375,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.007631368558761538,
|
| 153 |
+
"grad_norm": 5.826667785644531,
|
| 154 |
+
"learning_rate": 7.486009157642272e-07,
|
| 155 |
+
"loss": 4.356,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.007994767061559707,
|
| 160 |
+
"grad_norm": 6.022212982177734,
|
| 161 |
+
"learning_rate": 7.84940766044044e-07,
|
| 162 |
+
"loss": 4.2138,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.008358165564357875,
|
| 167 |
+
"grad_norm": 4.790489196777344,
|
| 168 |
+
"learning_rate": 8.212806163238608e-07,
|
| 169 |
+
"loss": 4.0662,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.008721564067156044,
|
| 174 |
+
"grad_norm": 4.448057174682617,
|
| 175 |
+
"learning_rate": 8.576204666036777e-07,
|
| 176 |
+
"loss": 3.9507,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.009084962569954213,
|
| 181 |
+
"grad_norm": 47.487003326416016,
|
| 182 |
+
"learning_rate": 8.939603168834945e-07,
|
| 183 |
+
"loss": 3.9576,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.009448361072752381,
|
| 188 |
+
"grad_norm": 8.64856243133545,
|
| 189 |
+
"learning_rate": 9.303001671633114e-07,
|
| 190 |
+
"loss": 3.7102,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.009811759575550548,
|
| 195 |
+
"grad_norm": 8.821709632873535,
|
| 196 |
+
"learning_rate": 9.66640017443128e-07,
|
| 197 |
+
"loss": 3.6644,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.010175158078348717,
|
| 202 |
+
"grad_norm": 14.071539878845215,
|
| 203 |
+
"learning_rate": 1.002979867722945e-06,
|
| 204 |
+
"loss": 3.6909,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.010538556581146885,
|
| 209 |
+
"grad_norm": 6.68039083480835,
|
| 210 |
+
"learning_rate": 1.0393197180027619e-06,
|
| 211 |
+
"loss": 3.6458,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.010901955083945054,
|
| 216 |
+
"grad_norm": 20.664649963378906,
|
| 217 |
+
"learning_rate": 1.0756595682825787e-06,
|
| 218 |
+
"loss": 3.6332,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.011265353586743222,
|
| 223 |
+
"grad_norm": 2.5272624492645264,
|
| 224 |
+
"learning_rate": 1.1119994185623955e-06,
|
| 225 |
+
"loss": 3.4835,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.01162875208954139,
|
| 230 |
+
"grad_norm": 8.353235244750977,
|
| 231 |
+
"learning_rate": 1.1483392688422123e-06,
|
| 232 |
+
"loss": 3.4892,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.01199215059233956,
|
| 237 |
+
"grad_norm": 7.0964531898498535,
|
| 238 |
+
"learning_rate": 1.1846791191220293e-06,
|
| 239 |
+
"loss": 3.5586,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.012355549095137728,
|
| 244 |
+
"grad_norm": 4.734161376953125,
|
| 245 |
+
"learning_rate": 1.2210189694018461e-06,
|
| 246 |
+
"loss": 3.4996,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.012718947597935897,
|
| 251 |
+
"grad_norm": 47.409996032714844,
|
| 252 |
+
"learning_rate": 1.257358819681663e-06,
|
| 253 |
+
"loss": 3.5565,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.013082346100734065,
|
| 258 |
+
"grad_norm": 2.880244016647339,
|
| 259 |
+
"learning_rate": 1.2936986699614797e-06,
|
| 260 |
+
"loss": 3.4154,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.013445744603532234,
|
| 265 |
+
"grad_norm": 6.637233734130859,
|
| 266 |
+
"learning_rate": 1.3300385202412968e-06,
|
| 267 |
+
"loss": 3.4119,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.013809143106330402,
|
| 272 |
+
"grad_norm": 11.791736602783203,
|
| 273 |
+
"learning_rate": 1.3663783705211136e-06,
|
| 274 |
+
"loss": 3.4347,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.01417254160912857,
|
| 279 |
+
"grad_norm": 8.274836540222168,
|
| 280 |
+
"learning_rate": 1.4027182208009304e-06,
|
| 281 |
+
"loss": 3.4253,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.01453594011192674,
|
| 286 |
+
"grad_norm": 10.09929084777832,
|
| 287 |
+
"learning_rate": 1.4390580710807472e-06,
|
| 288 |
+
"loss": 3.4702,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.014899338614724908,
|
| 293 |
+
"grad_norm": 6.32951545715332,
|
| 294 |
+
"learning_rate": 1.4753979213605642e-06,
|
| 295 |
+
"loss": 3.3513,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.015262737117523077,
|
| 300 |
+
"grad_norm": 2.4888486862182617,
|
| 301 |
+
"learning_rate": 1.511737771640381e-06,
|
| 302 |
+
"loss": 3.3421,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.015626135620321245,
|
| 307 |
+
"grad_norm": 3.02103328704834,
|
| 308 |
+
"learning_rate": 1.5480776219201978e-06,
|
| 309 |
+
"loss": 3.397,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.015989534123119414,
|
| 314 |
+
"grad_norm": 7.464268207550049,
|
| 315 |
+
"learning_rate": 1.5844174722000146e-06,
|
| 316 |
+
"loss": 3.3582,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.016352932625917582,
|
| 321 |
+
"grad_norm": 18.908123016357422,
|
| 322 |
+
"learning_rate": 1.6207573224798317e-06,
|
| 323 |
+
"loss": 3.4034,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.01671633112871575,
|
| 328 |
+
"grad_norm": 2.487326145172119,
|
| 329 |
+
"learning_rate": 1.6570971727596485e-06,
|
| 330 |
+
"loss": 3.2229,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.01707972963151392,
|
| 335 |
+
"grad_norm": 2.3999946117401123,
|
| 336 |
+
"learning_rate": 1.6934370230394653e-06,
|
| 337 |
+
"loss": 3.2185,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.017443128134312088,
|
| 342 |
+
"grad_norm": 5.007234573364258,
|
| 343 |
+
"learning_rate": 1.729776873319282e-06,
|
| 344 |
+
"loss": 3.2069,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.017806526637110257,
|
| 349 |
+
"grad_norm": 6.393301963806152,
|
| 350 |
+
"learning_rate": 1.766116723599099e-06,
|
| 351 |
+
"loss": 3.0687,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.018169925139908425,
|
| 356 |
+
"grad_norm": 45.44938278198242,
|
| 357 |
+
"learning_rate": 1.802456573878916e-06,
|
| 358 |
+
"loss": 3.112,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.018533323642706594,
|
| 363 |
+
"grad_norm": 7.32182502746582,
|
| 364 |
+
"learning_rate": 1.8387964241587327e-06,
|
| 365 |
+
"loss": 2.951,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.018896722145504762,
|
| 370 |
+
"grad_norm": 3.3864173889160156,
|
| 371 |
+
"learning_rate": 1.8751362744385495e-06,
|
| 372 |
+
"loss": 2.8879,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.019260120648302927,
|
| 377 |
+
"grad_norm": 5.429958343505859,
|
| 378 |
+
"learning_rate": 1.911476124718366e-06,
|
| 379 |
+
"loss": 2.7393,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.019623519151101096,
|
| 384 |
+
"grad_norm": 5.3577985763549805,
|
| 385 |
+
"learning_rate": 1.947815974998183e-06,
|
| 386 |
+
"loss": 2.4813,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.019986917653899264,
|
| 391 |
+
"grad_norm": 13.970659255981445,
|
| 392 |
+
"learning_rate": 1.9841558252779998e-06,
|
| 393 |
+
"loss": 2.3787,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.020350316156697433,
|
| 398 |
+
"grad_norm": 5.2666754722595215,
|
| 399 |
+
"learning_rate": 2.0204956755578166e-06,
|
| 400 |
+
"loss": 2.207,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.0207137146594956,
|
| 405 |
+
"grad_norm": 4.184991359710693,
|
| 406 |
+
"learning_rate": 2.0568355258376334e-06,
|
| 407 |
+
"loss": 2.0383,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.02107711316229377,
|
| 412 |
+
"grad_norm": 6.312343597412109,
|
| 413 |
+
"learning_rate": 2.09317537611745e-06,
|
| 414 |
+
"loss": 1.8416,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.02144051166509194,
|
| 419 |
+
"grad_norm": 4.754147529602051,
|
| 420 |
+
"learning_rate": 2.1295152263972674e-06,
|
| 421 |
+
"loss": 1.6002,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.021803910167890107,
|
| 426 |
+
"grad_norm": 21.47913360595703,
|
| 427 |
+
"learning_rate": 2.1658550766770842e-06,
|
| 428 |
+
"loss": 1.6015,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.021803910167890107,
|
| 433 |
+
"eval_loss": 1.5154471397399902,
|
| 434 |
+
"eval_runtime": 180.9184,
|
| 435 |
+
"eval_samples_per_second": 40.98,
|
| 436 |
+
"eval_steps_per_second": 5.124,
|
| 437 |
+
"eval_wer": 0.7997531177954872,
|
| 438 |
+
"step": 600
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.022167308670688276,
|
| 442 |
+
"grad_norm": 4.2374348640441895,
|
| 443 |
+
"learning_rate": 2.202194926956901e-06,
|
| 444 |
+
"loss": 1.4842,
|
| 445 |
+
"step": 610
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.022530707173486444,
|
| 449 |
+
"grad_norm": 4.392132759094238,
|
| 450 |
+
"learning_rate": 2.238534777236718e-06,
|
| 451 |
+
"loss": 1.3776,
|
| 452 |
+
"step": 620
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.022894105676284613,
|
| 456 |
+
"grad_norm": 4.682064533233643,
|
| 457 |
+
"learning_rate": 2.2748746275165347e-06,
|
| 458 |
+
"loss": 1.3177,
|
| 459 |
+
"step": 630
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.02325750417908278,
|
| 463 |
+
"grad_norm": 4.8396077156066895,
|
| 464 |
+
"learning_rate": 2.3112144777963515e-06,
|
| 465 |
+
"loss": 1.0737,
|
| 466 |
+
"step": 640
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.02362090268188095,
|
| 470 |
+
"grad_norm": 33.27382278442383,
|
| 471 |
+
"learning_rate": 2.3475543280761683e-06,
|
| 472 |
+
"loss": 1.3046,
|
| 473 |
+
"step": 650
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.02398430118467912,
|
| 477 |
+
"grad_norm": 5.410325050354004,
|
| 478 |
+
"learning_rate": 2.383894178355985e-06,
|
| 479 |
+
"loss": 1.1021,
|
| 480 |
+
"step": 660
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.024347699687477287,
|
| 484 |
+
"grad_norm": 3.9523680210113525,
|
| 485 |
+
"learning_rate": 2.420234028635802e-06,
|
| 486 |
+
"loss": 1.0602,
|
| 487 |
+
"step": 670
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.024711098190275456,
|
| 491 |
+
"grad_norm": 9.141073226928711,
|
| 492 |
+
"learning_rate": 2.456573878915619e-06,
|
| 493 |
+
"loss": 1.0631,
|
| 494 |
+
"step": 680
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.025074496693073624,
|
| 498 |
+
"grad_norm": 5.3534626960754395,
|
| 499 |
+
"learning_rate": 2.492913729195436e-06,
|
| 500 |
+
"loss": 0.8968,
|
| 501 |
+
"step": 690
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.025437895195871793,
|
| 505 |
+
"grad_norm": 32.30677795410156,
|
| 506 |
+
"learning_rate": 2.5292535794752527e-06,
|
| 507 |
+
"loss": 1.0439,
|
| 508 |
+
"step": 700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.02580129369866996,
|
| 512 |
+
"grad_norm": 4.310474872589111,
|
| 513 |
+
"learning_rate": 2.5655934297550696e-06,
|
| 514 |
+
"loss": 0.954,
|
| 515 |
+
"step": 710
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.02616469220146813,
|
| 519 |
+
"grad_norm": 5.586440563201904,
|
| 520 |
+
"learning_rate": 2.6019332800348864e-06,
|
| 521 |
+
"loss": 1.0031,
|
| 522 |
+
"step": 720
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.0265280907042663,
|
| 526 |
+
"grad_norm": 3.6927313804626465,
|
| 527 |
+
"learning_rate": 2.638273130314703e-06,
|
| 528 |
+
"loss": 0.7956,
|
| 529 |
+
"step": 730
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.026891489207064467,
|
| 533 |
+
"grad_norm": 4.270529747009277,
|
| 534 |
+
"learning_rate": 2.67461298059452e-06,
|
| 535 |
+
"loss": 0.8874,
|
| 536 |
+
"step": 740
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.027254887709862636,
|
| 540 |
+
"grad_norm": 23.553489685058594,
|
| 541 |
+
"learning_rate": 2.710952830874337e-06,
|
| 542 |
+
"loss": 0.8523,
|
| 543 |
+
"step": 750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.027618286212660804,
|
| 547 |
+
"grad_norm": 5.342041492462158,
|
| 548 |
+
"learning_rate": 2.747292681154154e-06,
|
| 549 |
+
"loss": 0.9029,
|
| 550 |
+
"step": 760
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.027981684715458973,
|
| 554 |
+
"grad_norm": 3.3802621364593506,
|
| 555 |
+
"learning_rate": 2.783632531433971e-06,
|
| 556 |
+
"loss": 0.8378,
|
| 557 |
+
"step": 770
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.02834508321825714,
|
| 561 |
+
"grad_norm": 6.378807067871094,
|
| 562 |
+
"learning_rate": 2.8199723817137876e-06,
|
| 563 |
+
"loss": 0.8085,
|
| 564 |
+
"step": 780
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.02870848172105531,
|
| 568 |
+
"grad_norm": 4.007000923156738,
|
| 569 |
+
"learning_rate": 2.8563122319936045e-06,
|
| 570 |
+
"loss": 0.8218,
|
| 571 |
+
"step": 790
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.02907188022385348,
|
| 575 |
+
"grad_norm": 68.16226196289062,
|
| 576 |
+
"learning_rate": 2.8926520822734213e-06,
|
| 577 |
+
"loss": 1.2055,
|
| 578 |
+
"step": 800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.029435278726651647,
|
| 582 |
+
"grad_norm": 6.70043420791626,
|
| 583 |
+
"learning_rate": 2.928991932553238e-06,
|
| 584 |
+
"loss": 0.7641,
|
| 585 |
+
"step": 810
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.029798677229449816,
|
| 589 |
+
"grad_norm": 5.498161315917969,
|
| 590 |
+
"learning_rate": 2.965331782833055e-06,
|
| 591 |
+
"loss": 0.7739,
|
| 592 |
+
"step": 820
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.030162075732247984,
|
| 596 |
+
"grad_norm": 9.515852928161621,
|
| 597 |
+
"learning_rate": 3.0016716331128717e-06,
|
| 598 |
+
"loss": 0.8293,
|
| 599 |
+
"step": 830
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.030525474235046153,
|
| 603 |
+
"grad_norm": 13.3881196975708,
|
| 604 |
+
"learning_rate": 3.0380114833926885e-06,
|
| 605 |
+
"loss": 0.5597,
|
| 606 |
+
"step": 840
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.03088887273784432,
|
| 610 |
+
"grad_norm": 13.670549392700195,
|
| 611 |
+
"learning_rate": 3.0743513336725057e-06,
|
| 612 |
+
"loss": 0.7658,
|
| 613 |
+
"step": 850
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.03125227124064249,
|
| 617 |
+
"grad_norm": 3.58305287361145,
|
| 618 |
+
"learning_rate": 3.1106911839523226e-06,
|
| 619 |
+
"loss": 0.7036,
|
| 620 |
+
"step": 860
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.031615669743440655,
|
| 624 |
+
"grad_norm": 4.119450569152832,
|
| 625 |
+
"learning_rate": 3.147031034232139e-06,
|
| 626 |
+
"loss": 0.6842,
|
| 627 |
+
"step": 870
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.03197906824623883,
|
| 631 |
+
"grad_norm": 6.412299156188965,
|
| 632 |
+
"learning_rate": 3.183370884511956e-06,
|
| 633 |
+
"loss": 0.7148,
|
| 634 |
+
"step": 880
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.03234246674903699,
|
| 638 |
+
"grad_norm": 8.700023651123047,
|
| 639 |
+
"learning_rate": 3.2197107347917726e-06,
|
| 640 |
+
"loss": 1.4861,
|
| 641 |
+
"step": 890
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.032705865251835164,
|
| 645 |
+
"grad_norm": 18.78075408935547,
|
| 646 |
+
"learning_rate": 3.25605058507159e-06,
|
| 647 |
+
"loss": 0.7162,
|
| 648 |
+
"step": 900
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.03306926375463333,
|
| 652 |
+
"grad_norm": 4.078335762023926,
|
| 653 |
+
"learning_rate": 3.292390435351406e-06,
|
| 654 |
+
"loss": 0.643,
|
| 655 |
+
"step": 910
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.0334326622574315,
|
| 659 |
+
"grad_norm": 6.603452682495117,
|
| 660 |
+
"learning_rate": 3.3287302856312234e-06,
|
| 661 |
+
"loss": 0.6623,
|
| 662 |
+
"step": 920
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.03379606076022967,
|
| 666 |
+
"grad_norm": 5.817732334136963,
|
| 667 |
+
"learning_rate": 3.3650701359110402e-06,
|
| 668 |
+
"loss": 0.6265,
|
| 669 |
+
"step": 930
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.03415945926302784,
|
| 673 |
+
"grad_norm": 8.310086250305176,
|
| 674 |
+
"learning_rate": 3.4014099861908575e-06,
|
| 675 |
+
"loss": 0.5343,
|
| 676 |
+
"step": 940
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.034522857765826004,
|
| 680 |
+
"grad_norm": NaN,
|
| 681 |
+
"learning_rate": 3.4341158514426923e-06,
|
| 682 |
+
"loss": 0.6999,
|
| 683 |
+
"step": 950
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.034886256268624176,
|
| 687 |
+
"grad_norm": 4.416926860809326,
|
| 688 |
+
"learning_rate": 3.4704557017225087e-06,
|
| 689 |
+
"loss": 0.6688,
|
| 690 |
+
"step": 960
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.03524965477142234,
|
| 694 |
+
"grad_norm": 3.2407495975494385,
|
| 695 |
+
"learning_rate": 3.506795552002326e-06,
|
| 696 |
+
"loss": 0.5808,
|
| 697 |
+
"step": 970
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.03561305327422051,
|
| 701 |
+
"grad_norm": NaN,
|
| 702 |
+
"learning_rate": 3.539501417254161e-06,
|
| 703 |
+
"loss": 3.0266,
|
| 704 |
+
"step": 980
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.03597645177701868,
|
| 708 |
+
"grad_norm": 8.086112022399902,
|
| 709 |
+
"learning_rate": 3.575841267533978e-06,
|
| 710 |
+
"loss": 0.5829,
|
| 711 |
+
"step": 990
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.03633985027981685,
|
| 715 |
+
"grad_norm": 252.45077514648438,
|
| 716 |
+
"learning_rate": 3.612181117813795e-06,
|
| 717 |
+
"loss": 0.7371,
|
| 718 |
+
"step": 1000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.036703248782615015,
|
| 722 |
+
"grad_norm": 3.5969936847686768,
|
| 723 |
+
"learning_rate": 3.648520968093612e-06,
|
| 724 |
+
"loss": 0.6632,
|
| 725 |
+
"step": 1010
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.03706664728541319,
|
| 729 |
+
"grad_norm": 3.0116841793060303,
|
| 730 |
+
"learning_rate": 3.6848608183734285e-06,
|
| 731 |
+
"loss": 0.537,
|
| 732 |
+
"step": 1020
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.03743004578821135,
|
| 736 |
+
"grad_norm": 5.494657039642334,
|
| 737 |
+
"learning_rate": 3.7212006686532457e-06,
|
| 738 |
+
"loss": 0.5422,
|
| 739 |
+
"step": 1030
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.037793444291009524,
|
| 743 |
+
"grad_norm": 21.526798248291016,
|
| 744 |
+
"learning_rate": 3.757540518933062e-06,
|
| 745 |
+
"loss": 0.5003,
|
| 746 |
+
"step": 1040
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.03815684279380769,
|
| 750 |
+
"grad_norm": 80.90055084228516,
|
| 751 |
+
"learning_rate": 3.7938803692128793e-06,
|
| 752 |
+
"loss": 0.6566,
|
| 753 |
+
"step": 1050
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.038520241296605855,
|
| 757 |
+
"grad_norm": 3.7678096294403076,
|
| 758 |
+
"learning_rate": 3.830220219492696e-06,
|
| 759 |
+
"loss": 0.5758,
|
| 760 |
+
"step": 1060
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.03888363979940403,
|
| 764 |
+
"grad_norm": 4.526616096496582,
|
| 765 |
+
"learning_rate": 3.866560069772512e-06,
|
| 766 |
+
"loss": 0.5648,
|
| 767 |
+
"step": 1070
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.03924703830220219,
|
| 771 |
+
"grad_norm": 4.571674346923828,
|
| 772 |
+
"learning_rate": 3.90289992005233e-06,
|
| 773 |
+
"loss": 0.5864,
|
| 774 |
+
"step": 1080
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.039610436805000364,
|
| 778 |
+
"grad_norm": 5.295219421386719,
|
| 779 |
+
"learning_rate": 3.939239770332146e-06,
|
| 780 |
+
"loss": 0.4476,
|
| 781 |
+
"step": 1090
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.03997383530779853,
|
| 785 |
+
"grad_norm": 16.631162643432617,
|
| 786 |
+
"learning_rate": 3.975579620611963e-06,
|
| 787 |
+
"loss": 0.6198,
|
| 788 |
+
"step": 1100
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.0403372338105967,
|
| 792 |
+
"grad_norm": 4.685397624969482,
|
| 793 |
+
"learning_rate": 4.01191947089178e-06,
|
| 794 |
+
"loss": 0.7512,
|
| 795 |
+
"step": 1110
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.040700632313394866,
|
| 799 |
+
"grad_norm": 3.333232879638672,
|
| 800 |
+
"learning_rate": 4.048259321171597e-06,
|
| 801 |
+
"loss": 0.5087,
|
| 802 |
+
"step": 1120
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.04106403081619304,
|
| 806 |
+
"grad_norm": 5.501911640167236,
|
| 807 |
+
"learning_rate": 4.084599171451414e-06,
|
| 808 |
+
"loss": 0.5772,
|
| 809 |
+
"step": 1130
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.0414274293189912,
|
| 813 |
+
"grad_norm": 8.066693305969238,
|
| 814 |
+
"learning_rate": 4.120939021731231e-06,
|
| 815 |
+
"loss": 0.4641,
|
| 816 |
+
"step": 1140
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.041790827821789375,
|
| 820 |
+
"grad_norm": 13.463829040527344,
|
| 821 |
+
"learning_rate": 4.1572788720110474e-06,
|
| 822 |
+
"loss": 0.5192,
|
| 823 |
+
"step": 1150
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.04215422632458754,
|
| 827 |
+
"grad_norm": 4.132773399353027,
|
| 828 |
+
"learning_rate": 4.193618722290864e-06,
|
| 829 |
+
"loss": 0.4696,
|
| 830 |
+
"step": 1160
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.04251762482738571,
|
| 834 |
+
"grad_norm": 6.176777362823486,
|
| 835 |
+
"learning_rate": 4.229958572570681e-06,
|
| 836 |
+
"loss": 0.4851,
|
| 837 |
+
"step": 1170
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.04288102333018388,
|
| 841 |
+
"grad_norm": 8.26610279083252,
|
| 842 |
+
"learning_rate": 4.266298422850498e-06,
|
| 843 |
+
"loss": 0.4967,
|
| 844 |
+
"step": 1180
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.04324442183298205,
|
| 848 |
+
"grad_norm": 3.9725544452667236,
|
| 849 |
+
"learning_rate": 4.302638273130315e-06,
|
| 850 |
+
"loss": 0.431,
|
| 851 |
+
"step": 1190
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.043607820335780215,
|
| 855 |
+
"grad_norm": 22.353294372558594,
|
| 856 |
+
"learning_rate": 4.338978123410132e-06,
|
| 857 |
+
"loss": 0.6523,
|
| 858 |
+
"step": 1200
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.043607820335780215,
|
| 862 |
+
"eval_loss": 0.5945897102355957,
|
| 863 |
+
"eval_runtime": 180.5168,
|
| 864 |
+
"eval_samples_per_second": 41.071,
|
| 865 |
+
"eval_steps_per_second": 5.135,
|
| 866 |
+
"eval_wer": 0.3718390908925881,
|
| 867 |
+
"step": 1200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.04397121883857839,
|
| 871 |
+
"grad_norm": 3.7954189777374268,
|
| 872 |
+
"learning_rate": 4.375317973689948e-06,
|
| 873 |
+
"loss": 0.4511,
|
| 874 |
+
"step": 1210
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.04433461734137655,
|
| 878 |
+
"grad_norm": 5.583435535430908,
|
| 879 |
+
"learning_rate": 4.411657823969766e-06,
|
| 880 |
+
"loss": 1.4019,
|
| 881 |
+
"step": 1220
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.044698015844174724,
|
| 885 |
+
"grad_norm": 8.544243812561035,
|
| 886 |
+
"learning_rate": 4.447997674249582e-06,
|
| 887 |
+
"loss": 0.4467,
|
| 888 |
+
"step": 1230
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.04506141434697289,
|
| 892 |
+
"grad_norm": 3.8716418743133545,
|
| 893 |
+
"learning_rate": 4.4843375245293996e-06,
|
| 894 |
+
"loss": 0.4347,
|
| 895 |
+
"step": 1240
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.04542481284977106,
|
| 899 |
+
"grad_norm": 19.459606170654297,
|
| 900 |
+
"learning_rate": 4.5206773748092155e-06,
|
| 901 |
+
"loss": 0.616,
|
| 902 |
+
"step": 1250
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.045788211352569226,
|
| 906 |
+
"grad_norm": 5.474793434143066,
|
| 907 |
+
"learning_rate": 4.557017225089033e-06,
|
| 908 |
+
"loss": 0.4689,
|
| 909 |
+
"step": 1260
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.0461516098553674,
|
| 913 |
+
"grad_norm": 4.705495834350586,
|
| 914 |
+
"learning_rate": 4.593357075368849e-06,
|
| 915 |
+
"loss": 0.4623,
|
| 916 |
+
"step": 1270
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.04651500835816556,
|
| 920 |
+
"grad_norm": 6.779942035675049,
|
| 921 |
+
"learning_rate": 4.629696925648667e-06,
|
| 922 |
+
"loss": 0.4418,
|
| 923 |
+
"step": 1280
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.046878406860963735,
|
| 927 |
+
"grad_norm": 6.802936553955078,
|
| 928 |
+
"learning_rate": 4.666036775928484e-06,
|
| 929 |
+
"loss": 0.4429,
|
| 930 |
+
"step": 1290
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.0472418053637619,
|
| 934 |
+
"grad_norm": 17.47754669189453,
|
| 935 |
+
"learning_rate": 4.7023766262083004e-06,
|
| 936 |
+
"loss": 0.596,
|
| 937 |
+
"step": 1300
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.04760520386656007,
|
| 941 |
+
"grad_norm": 4.036036968231201,
|
| 942 |
+
"learning_rate": 4.738716476488117e-06,
|
| 943 |
+
"loss": 0.4362,
|
| 944 |
+
"step": 1310
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.04796860236935824,
|
| 948 |
+
"grad_norm": 6.022701740264893,
|
| 949 |
+
"learning_rate": 4.775056326767934e-06,
|
| 950 |
+
"loss": 0.5092,
|
| 951 |
+
"step": 1320
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 0.04833200087215641,
|
| 955 |
+
"grad_norm": 5.533923625946045,
|
| 956 |
+
"learning_rate": 4.811396177047751e-06,
|
| 957 |
+
"loss": 0.4358,
|
| 958 |
+
"step": 1330
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.048695399374954575,
|
| 962 |
+
"grad_norm": 3.4037017822265625,
|
| 963 |
+
"learning_rate": 4.847736027327568e-06,
|
| 964 |
+
"loss": 0.3684,
|
| 965 |
+
"step": 1340
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.04905879787775275,
|
| 969 |
+
"grad_norm": 13.625974655151367,
|
| 970 |
+
"learning_rate": 4.8840758776073845e-06,
|
| 971 |
+
"loss": 0.583,
|
| 972 |
+
"step": 1350
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.04942219638055091,
|
| 976 |
+
"grad_norm": 3.597294330596924,
|
| 977 |
+
"learning_rate": 4.920415727887201e-06,
|
| 978 |
+
"loss": 0.4561,
|
| 979 |
+
"step": 1360
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.049785594883349084,
|
| 983 |
+
"grad_norm": 2.8846936225891113,
|
| 984 |
+
"learning_rate": 4.956755578167018e-06,
|
| 985 |
+
"loss": 0.409,
|
| 986 |
+
"step": 1370
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.05014899338614725,
|
| 990 |
+
"grad_norm": 5.500187397003174,
|
| 991 |
+
"learning_rate": 4.993095428446836e-06,
|
| 992 |
+
"loss": 0.4531,
|
| 993 |
+
"step": 1380
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.05051239188894542,
|
| 997 |
+
"grad_norm": 3.1203413009643555,
|
| 998 |
+
"learning_rate": 5.029435278726652e-06,
|
| 999 |
+
"loss": 0.4004,
|
| 1000 |
+
"step": 1390
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.050875790391743586,
|
| 1004 |
+
"grad_norm": 98.18115234375,
|
| 1005 |
+
"learning_rate": 5.0657751290064685e-06,
|
| 1006 |
+
"loss": 0.522,
|
| 1007 |
+
"step": 1400
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.05123918889454175,
|
| 1011 |
+
"grad_norm": 3.901418924331665,
|
| 1012 |
+
"learning_rate": 5.102114979286285e-06,
|
| 1013 |
+
"loss": 0.4041,
|
| 1014 |
+
"step": 1410
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.05160258739733992,
|
| 1018 |
+
"grad_norm": 4.045637130737305,
|
| 1019 |
+
"learning_rate": 5.138454829566102e-06,
|
| 1020 |
+
"loss": 0.4051,
|
| 1021 |
+
"step": 1420
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.05196598590013809,
|
| 1025 |
+
"grad_norm": 6.835183143615723,
|
| 1026 |
+
"learning_rate": 5.174794679845919e-06,
|
| 1027 |
+
"loss": 0.4937,
|
| 1028 |
+
"step": 1430
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.05232938440293626,
|
| 1032 |
+
"grad_norm": 7.708272457122803,
|
| 1033 |
+
"learning_rate": 5.211134530125736e-06,
|
| 1034 |
+
"loss": 0.3818,
|
| 1035 |
+
"step": 1440
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.052692782905734425,
|
| 1039 |
+
"grad_norm": 24.2607364654541,
|
| 1040 |
+
"learning_rate": 5.247474380405553e-06,
|
| 1041 |
+
"loss": 0.5445,
|
| 1042 |
+
"step": 1450
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.0530561814085326,
|
| 1046 |
+
"grad_norm": 3.3517005443573,
|
| 1047 |
+
"learning_rate": 5.283814230685369e-06,
|
| 1048 |
+
"loss": 0.4079,
|
| 1049 |
+
"step": 1460
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.05341957991133076,
|
| 1053 |
+
"grad_norm": 12.727778434753418,
|
| 1054 |
+
"learning_rate": 5.320154080965187e-06,
|
| 1055 |
+
"loss": 0.4285,
|
| 1056 |
+
"step": 1470
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.053782978414128935,
|
| 1060 |
+
"grad_norm": 4.984294891357422,
|
| 1061 |
+
"learning_rate": 5.356493931245003e-06,
|
| 1062 |
+
"loss": 0.5006,
|
| 1063 |
+
"step": 1480
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 0.0541463769169271,
|
| 1067 |
+
"grad_norm": 3.3041558265686035,
|
| 1068 |
+
"learning_rate": 5.392833781524821e-06,
|
| 1069 |
+
"loss": 0.3729,
|
| 1070 |
+
"step": 1490
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.05450977541972527,
|
| 1074 |
+
"grad_norm": 38.074546813964844,
|
| 1075 |
+
"learning_rate": 5.429173631804637e-06,
|
| 1076 |
+
"loss": 0.5401,
|
| 1077 |
+
"step": 1500
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.05487317392252344,
|
| 1081 |
+
"grad_norm": 5.649720668792725,
|
| 1082 |
+
"learning_rate": 5.465513482084454e-06,
|
| 1083 |
+
"loss": 0.3879,
|
| 1084 |
+
"step": 1510
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.05523657242532161,
|
| 1088 |
+
"grad_norm": 3.107583522796631,
|
| 1089 |
+
"learning_rate": 5.501853332364271e-06,
|
| 1090 |
+
"loss": 0.4144,
|
| 1091 |
+
"step": 1520
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.055599970928119774,
|
| 1095 |
+
"grad_norm": 19.246564865112305,
|
| 1096 |
+
"learning_rate": 5.538193182644088e-06,
|
| 1097 |
+
"loss": 0.4314,
|
| 1098 |
+
"step": 1530
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.055963369430917946,
|
| 1102 |
+
"grad_norm": 4.72367525100708,
|
| 1103 |
+
"learning_rate": 5.574533032923905e-06,
|
| 1104 |
+
"loss": 0.3576,
|
| 1105 |
+
"step": 1540
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.05632676793371611,
|
| 1109 |
+
"grad_norm": 25.88886260986328,
|
| 1110 |
+
"learning_rate": 5.6108728832037215e-06,
|
| 1111 |
+
"loss": 0.5385,
|
| 1112 |
+
"step": 1550
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.05669016643651428,
|
| 1116 |
+
"grad_norm": 3.1524956226348877,
|
| 1117 |
+
"learning_rate": 5.647212733483538e-06,
|
| 1118 |
+
"loss": 0.4075,
|
| 1119 |
+
"step": 1560
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.05705356493931245,
|
| 1123 |
+
"grad_norm": 3.883281707763672,
|
| 1124 |
+
"learning_rate": 5.683552583763355e-06,
|
| 1125 |
+
"loss": 0.4242,
|
| 1126 |
+
"step": 1570
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.05741696344211062,
|
| 1130 |
+
"grad_norm": 16.935935974121094,
|
| 1131 |
+
"learning_rate": 5.719892434043172e-06,
|
| 1132 |
+
"loss": 0.6194,
|
| 1133 |
+
"step": 1580
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.057780361944908785,
|
| 1137 |
+
"grad_norm": 4.23909330368042,
|
| 1138 |
+
"learning_rate": 5.756232284322989e-06,
|
| 1139 |
+
"loss": 0.4206,
|
| 1140 |
+
"step": 1590
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.05814376044770696,
|
| 1144 |
+
"grad_norm": 16.6039981842041,
|
| 1145 |
+
"learning_rate": 5.7925721346028056e-06,
|
| 1146 |
+
"loss": 0.4854,
|
| 1147 |
+
"step": 1600
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 0.05850715895050512,
|
| 1151 |
+
"grad_norm": 2.5220890045166016,
|
| 1152 |
+
"learning_rate": 5.828911984882622e-06,
|
| 1153 |
+
"loss": 0.4186,
|
| 1154 |
+
"step": 1610
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.058870557453303295,
|
| 1158 |
+
"grad_norm": 3.075101613998413,
|
| 1159 |
+
"learning_rate": 5.865251835162439e-06,
|
| 1160 |
+
"loss": 0.8877,
|
| 1161 |
+
"step": 1620
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.05923395595610146,
|
| 1165 |
+
"grad_norm": 5.511383056640625,
|
| 1166 |
+
"learning_rate": 5.901591685442257e-06,
|
| 1167 |
+
"loss": 0.4219,
|
| 1168 |
+
"step": 1630
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.05959735445889963,
|
| 1172 |
+
"grad_norm": 2.9449989795684814,
|
| 1173 |
+
"learning_rate": 5.937931535722073e-06,
|
| 1174 |
+
"loss": 0.2992,
|
| 1175 |
+
"step": 1640
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.0599607529616978,
|
| 1179 |
+
"grad_norm": 31.823612213134766,
|
| 1180 |
+
"learning_rate": 5.9742713860018905e-06,
|
| 1181 |
+
"loss": 0.5939,
|
| 1182 |
+
"step": 1650
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.06032415146449597,
|
| 1186 |
+
"grad_norm": 4.240995407104492,
|
| 1187 |
+
"learning_rate": 6.010611236281706e-06,
|
| 1188 |
+
"loss": 0.4176,
|
| 1189 |
+
"step": 1660
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.060687549967294134,
|
| 1193 |
+
"grad_norm": 2.6084980964660645,
|
| 1194 |
+
"learning_rate": 6.046951086561524e-06,
|
| 1195 |
+
"loss": 0.3542,
|
| 1196 |
+
"step": 1670
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.061050948470092306,
|
| 1200 |
+
"grad_norm": 8.318774223327637,
|
| 1201 |
+
"learning_rate": 6.08329093684134e-06,
|
| 1202 |
+
"loss": 0.3968,
|
| 1203 |
+
"step": 1680
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.06141434697289047,
|
| 1207 |
+
"grad_norm": 5.18604850769043,
|
| 1208 |
+
"learning_rate": 6.119630787121158e-06,
|
| 1209 |
+
"loss": 0.3879,
|
| 1210 |
+
"step": 1690
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.06177774547568864,
|
| 1214 |
+
"grad_norm": 51.732086181640625,
|
| 1215 |
+
"learning_rate": 6.1559706374009745e-06,
|
| 1216 |
+
"loss": 0.5025,
|
| 1217 |
+
"step": 1700
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.06214114397848681,
|
| 1221 |
+
"grad_norm": 2.5876500606536865,
|
| 1222 |
+
"learning_rate": 6.192310487680791e-06,
|
| 1223 |
+
"loss": 0.3558,
|
| 1224 |
+
"step": 1710
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.06250454248128498,
|
| 1228 |
+
"grad_norm": 5.071794033050537,
|
| 1229 |
+
"learning_rate": 6.228650337960608e-06,
|
| 1230 |
+
"loss": 0.3534,
|
| 1231 |
+
"step": 1720
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.06286794098408315,
|
| 1235 |
+
"grad_norm": 11.539891242980957,
|
| 1236 |
+
"learning_rate": 6.264990188240424e-06,
|
| 1237 |
+
"loss": 0.4628,
|
| 1238 |
+
"step": 1730
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.06323133948688131,
|
| 1242 |
+
"grad_norm": 3.275383710861206,
|
| 1243 |
+
"learning_rate": 6.301330038520243e-06,
|
| 1244 |
+
"loss": 0.3368,
|
| 1245 |
+
"step": 1740
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 0.06359473798967948,
|
| 1249 |
+
"grad_norm": 41.4942741394043,
|
| 1250 |
+
"learning_rate": 6.3376698888000586e-06,
|
| 1251 |
+
"loss": 0.53,
|
| 1252 |
+
"step": 1750
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 0.06395813649247765,
|
| 1256 |
+
"grad_norm": 3.0071399211883545,
|
| 1257 |
+
"learning_rate": 6.374009739079875e-06,
|
| 1258 |
+
"loss": 0.3623,
|
| 1259 |
+
"step": 1760
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.06432153499527582,
|
| 1263 |
+
"grad_norm": 3.385955333709717,
|
| 1264 |
+
"learning_rate": 6.410349589359692e-06,
|
| 1265 |
+
"loss": 0.3476,
|
| 1266 |
+
"step": 1770
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.06468493349807398,
|
| 1270 |
+
"grad_norm": 3.872527599334717,
|
| 1271 |
+
"learning_rate": 6.446689439639508e-06,
|
| 1272 |
+
"loss": 0.3168,
|
| 1273 |
+
"step": 1780
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.06504833200087215,
|
| 1277 |
+
"grad_norm": 4.668768882751465,
|
| 1278 |
+
"learning_rate": 6.483029289919327e-06,
|
| 1279 |
+
"loss": 0.3813,
|
| 1280 |
+
"step": 1790
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.06541173050367033,
|
| 1284 |
+
"grad_norm": 69.33656311035156,
|
| 1285 |
+
"learning_rate": 6.519369140199143e-06,
|
| 1286 |
+
"loss": 0.4557,
|
| 1287 |
+
"step": 1800
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.06541173050367033,
|
| 1291 |
+
"eval_loss": 0.5579342246055603,
|
| 1292 |
+
"eval_runtime": 180.31,
|
| 1293 |
+
"eval_samples_per_second": 41.118,
|
| 1294 |
+
"eval_steps_per_second": 5.141,
|
| 1295 |
+
"eval_wer": 0.34055221740156477,
|
| 1296 |
+
"step": 1800
|
| 1297 |
+
}
|
| 1298 |
+
],
|
| 1299 |
+
"logging_steps": 10,
|
| 1300 |
+
"max_steps": 165108,
|
| 1301 |
+
"num_input_tokens_seen": 0,
|
| 1302 |
+
"num_train_epochs": 6,
|
| 1303 |
+
"save_steps": 1800,
|
| 1304 |
+
"stateful_callbacks": {
|
| 1305 |
+
"TrainerControl": {
|
| 1306 |
+
"args": {
|
| 1307 |
+
"should_epoch_stop": false,
|
| 1308 |
+
"should_evaluate": false,
|
| 1309 |
+
"should_log": false,
|
| 1310 |
+
"should_save": true,
|
| 1311 |
+
"should_training_stop": false
|
| 1312 |
+
},
|
| 1313 |
+
"attributes": {}
|
| 1314 |
+
}
|
| 1315 |
+
},
|
| 1316 |
+
"total_flos": 5.891841119812055e+18,
|
| 1317 |
+
"train_batch_size": 16,
|
| 1318 |
+
"trial_name": null,
|
| 1319 |
+
"trial_params": null
|
| 1320 |
+
}
|
checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87933f1eb150a778b7fac52629f120e5917756e42ceb76d931ad6df3c999863b
|
| 3 |
+
size 5112
|