Upload 13 files

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "</s>": 35,
+  "<s>": 34
+}

config.json

@@ -0,0 +1,116 @@
+{
+  "activation_dropout": 0.0,
+  "adapter_attn_dim": null,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 768,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.075,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "output_hidden_size": 1024,
+  "pad_token_id": 28,
+  "proj_codevector_dim": 768,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.1",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 36,
+  "xvector_output_dim": 512
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4f552073354e252c0b04a1fb65677548710e26743d9f14838857e4868460c37
+size 1261955080

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee0261fb25bf9877515f6d7c7d203212440e28fe0d473b85d39671ab079258e5
+size 2490455431

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+    "do_normalize": true,
+    "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+    "feature_size": 1,
+    "padding_side": "right",
+    "padding_value": 0,
+    "return_attention_mask": true,
+    "sampling_rate": 16000
+}

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed8922833666f1d30991c244df891503c64b4888b5460d30f3394473e030a73b
+size 14709

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2da89ae85487ada19e0074af316009148c47a07e9bf2d70b87fed24b363c8c11
+size 1383

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2eab7e8b3a9d7500e08adcfa4f414a2516b4435cc6c481cfb352fc7d1c3f0764
+size 1721

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "[PAD]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,48 @@
+{
+  "added_tokens_decoder": {
+    "24": {
+      "content": "[UNK]",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "[PAD]",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "replace_word_delimiter_char": " ",
+  "target_lang": null,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "unk_token": "[UNK]",
+  "word_delimiter_token": "|"
+}

trainer_state.json

@@ -0,0 +1,249 @@
+{
+  "best_global_step": 1300,
+  "best_metric": 0.3449197860962567,
+  "best_model_checkpoint": "large-models/checkpoint-1300",
+  "epoch": 39.97841726618705,
+  "eval_steps": 100,
+  "global_step": 1360,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02877697841726619,
+      "grad_norm": NaN,
+      "learning_rate": 9.999999999999999e-06,
+      "loss": 95.1042,
+      "step": 1
+    },
+    {
+      "epoch": 2.920863309352518,
+      "grad_norm": 184.99476623535156,
+      "learning_rate": 3.900743895390339e-05,
+      "loss": 93.7411,
+      "step": 100
+    },
+    {
+      "epoch": 2.920863309352518,
+      "eval_cer": 0.910427807486631,
+      "eval_loss": 25.27887535095215,
+      "eval_runtime": 3.48,
+      "eval_samples_per_second": 34.77,
+      "eval_steps_per_second": 2.299,
+      "step": 100
+    },
+    {
+      "epoch": 5.863309352517986,
+      "grad_norm": 1.0720491409301758,
+      "learning_rate": 0.0001193619003949373,
+      "loss": 10.8676,
+      "step": 200
+    },
+    {
+      "epoch": 5.863309352517986,
+      "eval_cer": 0.910427807486631,
+      "eval_loss": 3.770867109298706,
+      "eval_runtime": 3.4378,
+      "eval_samples_per_second": 35.197,
+      "eval_steps_per_second": 2.327,
+      "step": 200
+    },
+    {
+      "epoch": 8.805755395683454,
+      "grad_norm": 1.829892635345459,
+      "learning_rate": 0.00020574617217952577,
+      "loss": 3.7416,
+      "step": 300
+    },
+    {
+      "epoch": 8.805755395683454,
+      "eval_cer": 0.910427807486631,
+      "eval_loss": 3.518134832382202,
+      "eval_runtime": 3.4621,
+      "eval_samples_per_second": 34.95,
+      "eval_steps_per_second": 2.311,
+      "step": 300
+    },
+    {
+      "epoch": 11.748201438848922,
+      "grad_norm": 2.050036668777466,
+      "learning_rate": 0.00024919654886546277,
+      "loss": 3.5984,
+      "step": 400
+    },
+    {
+      "epoch": 11.748201438848922,
+      "eval_cer": 0.910427807486631,
+      "eval_loss": 3.443953275680542,
+      "eval_runtime": 3.3861,
+      "eval_samples_per_second": 35.735,
+      "eval_steps_per_second": 2.363,
+      "step": 400
+    },
+    {
+      "epoch": 14.690647482014388,
+      "grad_norm": 1.6520733833312988,
+      "learning_rate": 0.0002451146847844765,
+      "loss": 3.4592,
+      "step": 500
+    },
+    {
+      "epoch": 14.690647482014388,
+      "eval_cer": 0.910427807486631,
+      "eval_loss": 3.0703835487365723,
+      "eval_runtime": 3.3714,
+      "eval_samples_per_second": 35.89,
+      "eval_steps_per_second": 2.373,
+      "step": 500
+    },
+    {
+      "epoch": 17.633093525179856,
+      "grad_norm": 1.6168930530548096,
+      "learning_rate": 0.00022742037782978917,
+      "loss": 2.8283,
+      "step": 600
+    },
+    {
+      "epoch": 17.633093525179856,
+      "eval_cer": 0.5975935828877005,
+      "eval_loss": 1.7633609771728516,
+      "eval_runtime": 3.383,
+      "eval_samples_per_second": 35.767,
+      "eval_steps_per_second": 2.365,
+      "step": 600
+    },
+    {
+      "epoch": 20.575539568345324,
+      "grad_norm": 1.3846379518508911,
+      "learning_rate": 0.0001986734489292734,
+      "loss": 1.6807,
+      "step": 700
+    },
+    {
+      "epoch": 20.575539568345324,
+      "eval_cer": 0.5013368983957219,
+      "eval_loss": 1.0697791576385498,
+      "eval_runtime": 3.3782,
+      "eval_samples_per_second": 35.818,
+      "eval_steps_per_second": 2.368,
+      "step": 700
+    },
+    {
+      "epoch": 23.51798561151079,
+      "grad_norm": 3.318430185317993,
+      "learning_rate": 0.00016197611753459705,
+      "loss": 1.1373,
+      "step": 800
+    },
+    {
+      "epoch": 23.51798561151079,
+      "eval_cer": 0.40641711229946526,
+      "eval_loss": 0.9486581087112427,
+      "eval_runtime": 3.443,
+      "eval_samples_per_second": 35.143,
+      "eval_steps_per_second": 2.324,
+      "step": 800
+    },
+    {
+      "epoch": 26.46043165467626,
+      "grad_norm": 1.7900818586349487,
+      "learning_rate": 0.00012128856929448933,
+      "loss": 0.8964,
+      "step": 900
+    },
+    {
+      "epoch": 26.46043165467626,
+      "eval_cer": 0.37566844919786097,
+      "eval_loss": 0.9327743649482727,
+      "eval_runtime": 3.3762,
+      "eval_samples_per_second": 35.839,
+      "eval_steps_per_second": 2.37,
+      "step": 900
+    },
+    {
+      "epoch": 29.402877697841728,
+      "grad_norm": 2.1195766925811768,
+      "learning_rate": 8.100159336169578e-05,
+      "loss": 0.7525,
+      "step": 1000
+    },
+    {
+      "epoch": 29.402877697841728,
+      "eval_cer": 0.3609625668449198,
+      "eval_loss": 0.8588594198226929,
+      "eval_runtime": 3.3352,
+      "eval_samples_per_second": 36.28,
+      "eval_steps_per_second": 2.399,
+      "step": 1000
+    },
+    {
+      "epoch": 32.34532374100719,
+      "grad_norm": 2.070112705230713,
+      "learning_rate": 4.5462751203394424e-05,
+      "loss": 0.6327,
+      "step": 1100
+    },
+    {
+      "epoch": 32.34532374100719,
+      "eval_cer": 0.35561497326203206,
+      "eval_loss": 0.8075717091560364,
+      "eval_runtime": 3.3241,
+      "eval_samples_per_second": 36.401,
+      "eval_steps_per_second": 2.407,
+      "step": 1100
+    },
+    {
+      "epoch": 35.28776978417266,
+      "grad_norm": 2.222806692123413,
+      "learning_rate": 1.850721031920824e-05,
+      "loss": 0.5846,
+      "step": 1200
+    },
+    {
+      "epoch": 35.28776978417266,
+      "eval_cer": 0.37433155080213903,
+      "eval_loss": 0.8168951272964478,
+      "eval_runtime": 3.443,
+      "eval_samples_per_second": 35.144,
+      "eval_steps_per_second": 2.324,
+      "step": 1200
+    },
+    {
+      "epoch": 38.23021582733813,
+      "grad_norm": 2.681779146194458,
+      "learning_rate": 3.0438728583211972e-06,
+      "loss": 0.5549,
+      "step": 1300
+    },
+    {
+      "epoch": 38.23021582733813,
+      "eval_cer": 0.3449197860962567,
+      "eval_loss": 0.7855556607246399,
+      "eval_runtime": 3.4484,
+      "eval_samples_per_second": 35.089,
+      "eval_steps_per_second": 2.32,
+      "step": 1300
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 1360,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 40,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.085833579033526e+19,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3bf28e3573de66491b6342572979a11f66ce2a082c7efec06d1fbc5c7bbdfcb
+size 5713

vocab.json

@@ -0,0 +1,36 @@
+{
+  "[PAD]": 28,
+  "[UNK]": 24,
+  "a": 0,
+  "b": 16,
+  "c": 17,
+  "d": 18,
+  "e": 11,
+  "f": 25,
+  "g": 27,
+  "h": 30,
+  "i": 8,
+  "j": 32,
+  "k": 21,
+  "l": 3,
+  "m": 26,
+  "n": 15,
+  "o": 1,
+  "p": 13,
+  "r": 2,
+  "s": 20,
+  "t": 33,
+  "u": 4,
+  "v": 19,
+  "y": 31,
+  "z": 5,
+  "ą": 23,
+  "č": 9,
+  "ė": 12,
+  "ę": 7,
+  "į": 14,
+  "š": 22,
+  "ū": 29,
+  "ų": 10,
+  "ž": 6
+}