eustlb HF Staff commited on
Commit
202b67b
·
verified ·
1 Parent(s): 1454393

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -1,3 +1,150 @@
1
- ---
2
- license: cc-by-4.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ language:
4
+ - en
5
+ library_name: moshi
6
+ tags:
7
+ - audio
8
+ - automatic-speech-recognition
9
+ ---
10
+ # Model Card for Kyutai STT
11
+
12
+ This repo is meant to use the model with [Transformers](https://github.com/huggingface/transformers) 🤗
13
+
14
+ Install Transformers from source:
15
+ ```bash
16
+ pip install git+https://github.com/huggingface/transformers
17
+ ```
18
+
19
+ Inference:
20
+ ```python
21
+ import torch
22
+ from datasets import load_dataset, Audio
23
+ from transformers import KyutaiSpeechToTextProcessor, KyutaiSpeechToTextForConditionalGeneration
24
+
25
+ # 1. load the model and the processor
26
+ torch_device = "cuda" if torch.cuda.is_available() else "cpu"
27
+ model_id = "kyutai/stt-2.6b-en-trfs"
28
+
29
+ processor = KyutaiSpeechToTextProcessor.from_pretrained(model_id)
30
+ model = KyutaiSpeechToTextForConditionalGeneration.from_pretrained(model_id, device_map=torch_device)
31
+
32
+ # 2. load audio samples
33
+ ds = load_dataset(
34
+ "hf-internal-testing/librispeech_asr_dummy", "clean", split="validation"
35
+ )
36
+ ds = ds.cast_column("audio", Audio(sampling_rate=24000))
37
+
38
+ # 3. prepare the model inputs
39
+ inputs = processor(
40
+ ds[0]["audio"]["array"],
41
+ )
42
+ inputs.to(torch_device)
43
+
44
+ # 4. infer the model
45
+ output_tokens = model.generate(**inputs)
46
+
47
+ # 5. decode the generated tokens
48
+ print(processor.batch_decode(output_tokens, skip_special_tokens=True))
49
+ ```
50
+
51
+ Batched inference:
52
+ ```python
53
+ import torch
54
+ from datasets import load_dataset, Audio
55
+ from transformers import KyutaiSpeechToTextProcessor, KyutaiSpeechToTextForConditionalGeneration
56
+
57
+ # 1. load the model and the processor
58
+ torch_device = "cuda" if torch.cuda.is_available() else "cpu"
59
+ model_id = "kyutai/stt-2.6b-en-trfs"
60
+
61
+ processor = KyutaiSpeechToTextProcessor.from_pretrained(model_id)
62
+ model = KyutaiSpeechToTextForConditionalGeneration.from_pretrained(model_id, device_map=torch_device)
63
+
64
+ # 2. load audio samples
65
+ ds = load_dataset(
66
+ "hf-internal-testing/librispeech_asr_dummy", "clean", split="validation"
67
+ )
68
+ ds = ds.cast_column("audio", Audio(sampling_rate=24000))
69
+
70
+ # 3. prepare the model inputs
71
+ audio_arrays = [ds[i]["audio"]["array"] for i in range(4)]
72
+ inputs = processor(audio_arrays, return_tensors="pt", padding=True)
73
+ inputs = inputs.to(torch_device)
74
+
75
+ # 4. infer the model
76
+ output_tokens = model.generate(**inputs)
77
+
78
+ # 5. decode the generated tokens
79
+ decoded_outputs = processor.batch_decode(output_tokens, skip_special_tokens=True)
80
+ for output in decoded_outputs:
81
+ print(output)
82
+ ```
83
+
84
+ See also the [project page](https://kyutai.org/next/stt)
85
+ and the [GitHub repository](https://github.com/kyutai-labs/delayed-streams-modeling/).
86
+
87
+ This is a model for streaming speech-to-text (STT, also known as automatic speech recognition, ASR).
88
+ Unlike offline speech-to-text, where the model needs the entire audio to produce the transcript,
89
+ our model starts to output the transcript as soon as a few seconds of audio become available.
90
+
91
+ ## Model Details
92
+
93
+ The model architecture is a Transformer that consumes audio tokenized by Mimi (see [the Moshi paper](https://arxiv.org/abs/2410.00037)) and outputs text tokens.
94
+ The frame rate is 12.5 Hz and each audio frame is represented by 32 audio tokens.
95
+
96
+ We release two models:
97
+ - `kyutai/stt-1b-en_fr`, an English and French model with ~1B parameters, a 0.5 second delay, and a [semantic VAD](https://kyutai.org/next/stt#semantic-vad).
98
+ - `kyutai/stt-2.6b-en`, an English-only model with ~2.6B parameters and a 2.5 second delay.
99
+
100
+ ## Model Description
101
+
102
+ Kyutai STT is a decoder-only model for streaming speech-to-text.
103
+ It leverages the multistream architecture of [Moshi](https://moshi.chat/) to model text stream based on the speech stream.
104
+ The text stream is shifted w.r.t. the audio stream to allow the model to predict text tokens based on the input audio.
105
+
106
+ * Developed by: Kyutai
107
+ * Model type: Streaming Speech-to-Text transcription.
108
+ * Language(s) (NLP): English and French for `kyutai/stt-1b-en_fr`, English for `kyutai/stt-2.6b-en`
109
+ * License: Model weights are licensed under CC-BY 4.0
110
+ * Repository: [GitHub](https://github.com/kyutai-labs/delayed-streams-modeling/)
111
+
112
+ ## Uses
113
+
114
+ ### Direct Use
115
+
116
+ The model can be used for streaming speech-to-text.
117
+ It is robust to noisy conditions and was found to perform well with audio as long as 2 hours with no additonal changes.
118
+ The model produces transcripts with capitalization and punctuation.
119
+ The predicted text token timestamps can be recovered by subtracting the model's text stream offset (0.5 or 2.5 seconds) from the frame's offset.
120
+
121
+ ## How to Get Started with the Model
122
+
123
+ See the [GitHub repository](https://github.com/kyutai-labs/delayed-streams-modeling/).
124
+
125
+ ## Training Details
126
+
127
+ ### Training Data
128
+
129
+ Pretraining stage: For both `kyutai/stt-2.6b-en` and `kyutai/stt-1b-en_fr`, we use an audio collection of 2.5 million hours of publicly available audio content.
130
+ For this dataset, we obtained synthetic transcripts by running [whisper-timestamped](https://github.com/linto-ai/whisper-timestamped).
131
+
132
+ For `kyutai/stt-2.6b-en`:
133
+
134
+ - Finetuning stage: We then finetune the model on a collection of public datasets with
135
+ ground-truth transcripts. This dataset contains 24000 hours of audio.
136
+
137
+ - Long-form finetuning stage: Finally, we finetune the model on a combination of data from the previous stage and long-form audio.
138
+ The long-form audio is obtained from two sources: (a) concatenating LibriSpeech examples (1000 hours), (b) synthesizing dialogs (22000 hours).
139
+
140
+ For `kyutai/stt-1b-en_fr`:
141
+
142
+ - Finetuning stage: We finetune on the Fisher dataset of 2000 hours of English audio, plus proprietary data (1000 hours in English, 600 hours in French).
143
+
144
+ ### Compute Infrastructure
145
+
146
+ Pretraining and finetuning was done with 48 and 16 H100 Nvidia GPUs, respectively.
147
+
148
+ ## Model Card Authors
149
+
150
+ Neil Zeghidour, Eugene Kharitonov, Manu Orsini, Václav Volhejn, Gabriel de Marmiesse, Edouard Grave, Patrick Perez, Laurent Mazaré, Alexandre Défossez
config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "KyutaiSpeechToTextForConditionalGeneration"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "audio_bos_token_id": 2048,
7
+ "audio_pad_token_id": 69569,
8
+ "bos_token_id": 48000,
9
+ "codebook_vocab_size": 2049,
10
+ "codec_config": {
11
+ "_frame_rate": null,
12
+ "attention_bias": false,
13
+ "attention_dropout": 0.0,
14
+ "audio_channels": 1,
15
+ "codebook_dim": 256,
16
+ "codebook_size": 2048,
17
+ "compress": 2,
18
+ "dilation_growth_rate": 2,
19
+ "head_dim": 64,
20
+ "hidden_act": "gelu",
21
+ "hidden_size": 512,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 2048,
24
+ "kernel_size": 7,
25
+ "last_kernel_size": 3,
26
+ "layer_scale_initial_scale": 0.01,
27
+ "max_position_embeddings": 8000,
28
+ "model_type": "mimi",
29
+ "norm_eps": 1e-05,
30
+ "num_attention_heads": 8,
31
+ "num_filters": 64,
32
+ "num_hidden_layers": 8,
33
+ "num_key_value_heads": 8,
34
+ "num_quantizers": 32,
35
+ "num_residual_layers": 1,
36
+ "num_semantic_quantizers": 1,
37
+ "pad_mode": "constant",
38
+ "residual_kernel_size": 3,
39
+ "rope_theta": 10000.0,
40
+ "sampling_rate": 24000,
41
+ "sliding_window": 250,
42
+ "trim_right_ratio": 1.0,
43
+ "upsample_groups": 512,
44
+ "upsampling_ratios": [
45
+ 8,
46
+ 6,
47
+ 5,
48
+ 4
49
+ ],
50
+ "use_cache": false,
51
+ "use_causal_conv": true,
52
+ "use_conv_shortcut": false,
53
+ "use_streaming": false,
54
+ "vector_quantization_hidden_dimension": 256
55
+ },
56
+ "ffn_dim": 11264,
57
+ "frame_size": 1920,
58
+ "head_dim": 64,
59
+ "hidden_act": "silu",
60
+ "hidden_size": 2048,
61
+ "initializer_range": 0.02,
62
+ "max_position_embeddings": 750,
63
+ "model_type": "kyutai_speech_to_text",
64
+ "num_attention_heads": 32,
65
+ "num_codebooks": 32,
66
+ "num_hidden_layers": 48,
67
+ "num_key_value_heads": 32,
68
+ "pad_token_id": 3,
69
+ "rms_norm_eps": 1e-08,
70
+ "rope_theta": 100000.0,
71
+ "sliding_window": 375,
72
+ "tie_word_embeddings": false,
73
+ "torch_dtype": "float32",
74
+ "transformers_version": "4.53.0.dev0",
75
+ "use_cache": true,
76
+ "vocab_size": 4001
77
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_window_size": 1,
3
+ "bos_token_id": 48000,
4
+ "cache_implementation": "sliding_window",
5
+ "codec_cache_implementation": "sliding_window",
6
+ "codec_use_cache": true,
7
+ "pad_token_id": 3,
8
+ "transformers_version": "4.53.0.dev0"
9
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d10da996924f8718e69ad188d65e4334f6c82a12bcaf8c5df83e80c9f5fca98b
3
+ size 4372392344
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe01b3882f795cc9198c6e7f1fa4f08000e14f59ac8b181a294535ddb41cc33b
3
+ size 1547899924
model.safetensors.index.json ADDED
@@ -0,0 +1,745 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 2696431425,
4
+ "total_size": 6759058820
5
+ },
6
+ "weight_map": {
7
+ "codec_model.decoder.layers.0.conv.bias": "model-00002-of-00002.safetensors",
8
+ "codec_model.decoder.layers.0.conv.weight": "model-00002-of-00002.safetensors",
9
+ "codec_model.decoder.layers.11.conv.bias": "model-00002-of-00002.safetensors",
10
+ "codec_model.decoder.layers.11.conv.weight": "model-00002-of-00002.safetensors",
11
+ "codec_model.decoder.layers.12.block.1.conv.bias": "model-00002-of-00002.safetensors",
12
+ "codec_model.decoder.layers.12.block.1.conv.weight": "model-00002-of-00002.safetensors",
13
+ "codec_model.decoder.layers.12.block.3.conv.bias": "model-00002-of-00002.safetensors",
14
+ "codec_model.decoder.layers.12.block.3.conv.weight": "model-00002-of-00002.safetensors",
15
+ "codec_model.decoder.layers.14.conv.bias": "model-00002-of-00002.safetensors",
16
+ "codec_model.decoder.layers.14.conv.weight": "model-00002-of-00002.safetensors",
17
+ "codec_model.decoder.layers.2.conv.bias": "model-00002-of-00002.safetensors",
18
+ "codec_model.decoder.layers.2.conv.weight": "model-00002-of-00002.safetensors",
19
+ "codec_model.decoder.layers.3.block.1.conv.bias": "model-00002-of-00002.safetensors",
20
+ "codec_model.decoder.layers.3.block.1.conv.weight": "model-00002-of-00002.safetensors",
21
+ "codec_model.decoder.layers.3.block.3.conv.bias": "model-00002-of-00002.safetensors",
22
+ "codec_model.decoder.layers.3.block.3.conv.weight": "model-00002-of-00002.safetensors",
23
+ "codec_model.decoder.layers.5.conv.bias": "model-00002-of-00002.safetensors",
24
+ "codec_model.decoder.layers.5.conv.weight": "model-00002-of-00002.safetensors",
25
+ "codec_model.decoder.layers.6.block.1.conv.bias": "model-00002-of-00002.safetensors",
26
+ "codec_model.decoder.layers.6.block.1.conv.weight": "model-00002-of-00002.safetensors",
27
+ "codec_model.decoder.layers.6.block.3.conv.bias": "model-00002-of-00002.safetensors",
28
+ "codec_model.decoder.layers.6.block.3.conv.weight": "model-00002-of-00002.safetensors",
29
+ "codec_model.decoder.layers.8.conv.bias": "model-00002-of-00002.safetensors",
30
+ "codec_model.decoder.layers.8.conv.weight": "model-00002-of-00002.safetensors",
31
+ "codec_model.decoder.layers.9.block.1.conv.bias": "model-00002-of-00002.safetensors",
32
+ "codec_model.decoder.layers.9.block.1.conv.weight": "model-00002-of-00002.safetensors",
33
+ "codec_model.decoder.layers.9.block.3.conv.bias": "model-00002-of-00002.safetensors",
34
+ "codec_model.decoder.layers.9.block.3.conv.weight": "model-00002-of-00002.safetensors",
35
+ "codec_model.decoder_transformer.layers.0.input_layernorm.bias": "model-00002-of-00002.safetensors",
36
+ "codec_model.decoder_transformer.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
37
+ "codec_model.decoder_transformer.layers.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
38
+ "codec_model.decoder_transformer.layers.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
39
+ "codec_model.decoder_transformer.layers.0.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
40
+ "codec_model.decoder_transformer.layers.0.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
41
+ "codec_model.decoder_transformer.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
42
+ "codec_model.decoder_transformer.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
43
+ "codec_model.decoder_transformer.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
44
+ "codec_model.decoder_transformer.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
45
+ "codec_model.decoder_transformer.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
46
+ "codec_model.decoder_transformer.layers.0.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
47
+ "codec_model.decoder_transformer.layers.1.input_layernorm.bias": "model-00002-of-00002.safetensors",
48
+ "codec_model.decoder_transformer.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
49
+ "codec_model.decoder_transformer.layers.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
50
+ "codec_model.decoder_transformer.layers.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
51
+ "codec_model.decoder_transformer.layers.1.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
52
+ "codec_model.decoder_transformer.layers.1.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
53
+ "codec_model.decoder_transformer.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
54
+ "codec_model.decoder_transformer.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
55
+ "codec_model.decoder_transformer.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
56
+ "codec_model.decoder_transformer.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
57
+ "codec_model.decoder_transformer.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
58
+ "codec_model.decoder_transformer.layers.1.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
59
+ "codec_model.decoder_transformer.layers.2.input_layernorm.bias": "model-00002-of-00002.safetensors",
60
+ "codec_model.decoder_transformer.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
61
+ "codec_model.decoder_transformer.layers.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
62
+ "codec_model.decoder_transformer.layers.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
63
+ "codec_model.decoder_transformer.layers.2.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
64
+ "codec_model.decoder_transformer.layers.2.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
65
+ "codec_model.decoder_transformer.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
66
+ "codec_model.decoder_transformer.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
67
+ "codec_model.decoder_transformer.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
68
+ "codec_model.decoder_transformer.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
69
+ "codec_model.decoder_transformer.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
70
+ "codec_model.decoder_transformer.layers.2.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
71
+ "codec_model.decoder_transformer.layers.3.input_layernorm.bias": "model-00002-of-00002.safetensors",
72
+ "codec_model.decoder_transformer.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
73
+ "codec_model.decoder_transformer.layers.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
74
+ "codec_model.decoder_transformer.layers.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
75
+ "codec_model.decoder_transformer.layers.3.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
76
+ "codec_model.decoder_transformer.layers.3.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
77
+ "codec_model.decoder_transformer.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
78
+ "codec_model.decoder_transformer.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
79
+ "codec_model.decoder_transformer.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
80
+ "codec_model.decoder_transformer.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
81
+ "codec_model.decoder_transformer.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
82
+ "codec_model.decoder_transformer.layers.3.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
83
+ "codec_model.decoder_transformer.layers.4.input_layernorm.bias": "model-00002-of-00002.safetensors",
84
+ "codec_model.decoder_transformer.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors",
85
+ "codec_model.decoder_transformer.layers.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
86
+ "codec_model.decoder_transformer.layers.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
87
+ "codec_model.decoder_transformer.layers.4.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
88
+ "codec_model.decoder_transformer.layers.4.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
89
+ "codec_model.decoder_transformer.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
90
+ "codec_model.decoder_transformer.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
91
+ "codec_model.decoder_transformer.layers.4.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
92
+ "codec_model.decoder_transformer.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
93
+ "codec_model.decoder_transformer.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
94
+ "codec_model.decoder_transformer.layers.4.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
95
+ "codec_model.decoder_transformer.layers.5.input_layernorm.bias": "model-00002-of-00002.safetensors",
96
+ "codec_model.decoder_transformer.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors",
97
+ "codec_model.decoder_transformer.layers.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
98
+ "codec_model.decoder_transformer.layers.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
99
+ "codec_model.decoder_transformer.layers.5.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
100
+ "codec_model.decoder_transformer.layers.5.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
101
+ "codec_model.decoder_transformer.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
102
+ "codec_model.decoder_transformer.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
103
+ "codec_model.decoder_transformer.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
104
+ "codec_model.decoder_transformer.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
105
+ "codec_model.decoder_transformer.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
106
+ "codec_model.decoder_transformer.layers.5.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
107
+ "codec_model.decoder_transformer.layers.6.input_layernorm.bias": "model-00002-of-00002.safetensors",
108
+ "codec_model.decoder_transformer.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors",
109
+ "codec_model.decoder_transformer.layers.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
110
+ "codec_model.decoder_transformer.layers.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
111
+ "codec_model.decoder_transformer.layers.6.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
112
+ "codec_model.decoder_transformer.layers.6.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
113
+ "codec_model.decoder_transformer.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
114
+ "codec_model.decoder_transformer.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
115
+ "codec_model.decoder_transformer.layers.6.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
116
+ "codec_model.decoder_transformer.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
117
+ "codec_model.decoder_transformer.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
118
+ "codec_model.decoder_transformer.layers.6.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
119
+ "codec_model.decoder_transformer.layers.7.input_layernorm.bias": "model-00002-of-00002.safetensors",
120
+ "codec_model.decoder_transformer.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors",
121
+ "codec_model.decoder_transformer.layers.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
122
+ "codec_model.decoder_transformer.layers.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
123
+ "codec_model.decoder_transformer.layers.7.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
124
+ "codec_model.decoder_transformer.layers.7.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
125
+ "codec_model.decoder_transformer.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
126
+ "codec_model.decoder_transformer.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
127
+ "codec_model.decoder_transformer.layers.7.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
128
+ "codec_model.decoder_transformer.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
129
+ "codec_model.decoder_transformer.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
130
+ "codec_model.decoder_transformer.layers.7.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
131
+ "codec_model.downsample.conv.weight": "model-00002-of-00002.safetensors",
132
+ "codec_model.encoder.layers.0.conv.bias": "model-00002-of-00002.safetensors",
133
+ "codec_model.encoder.layers.0.conv.weight": "model-00002-of-00002.safetensors",
134
+ "codec_model.encoder.layers.1.block.1.conv.bias": "model-00002-of-00002.safetensors",
135
+ "codec_model.encoder.layers.1.block.1.conv.weight": "model-00002-of-00002.safetensors",
136
+ "codec_model.encoder.layers.1.block.3.conv.bias": "model-00002-of-00002.safetensors",
137
+ "codec_model.encoder.layers.1.block.3.conv.weight": "model-00002-of-00002.safetensors",
138
+ "codec_model.encoder.layers.10.block.1.conv.bias": "model-00002-of-00002.safetensors",
139
+ "codec_model.encoder.layers.10.block.1.conv.weight": "model-00002-of-00002.safetensors",
140
+ "codec_model.encoder.layers.10.block.3.conv.bias": "model-00002-of-00002.safetensors",
141
+ "codec_model.encoder.layers.10.block.3.conv.weight": "model-00002-of-00002.safetensors",
142
+ "codec_model.encoder.layers.12.conv.bias": "model-00002-of-00002.safetensors",
143
+ "codec_model.encoder.layers.12.conv.weight": "model-00002-of-00002.safetensors",
144
+ "codec_model.encoder.layers.14.conv.bias": "model-00002-of-00002.safetensors",
145
+ "codec_model.encoder.layers.14.conv.weight": "model-00002-of-00002.safetensors",
146
+ "codec_model.encoder.layers.3.conv.bias": "model-00002-of-00002.safetensors",
147
+ "codec_model.encoder.layers.3.conv.weight": "model-00002-of-00002.safetensors",
148
+ "codec_model.encoder.layers.4.block.1.conv.bias": "model-00002-of-00002.safetensors",
149
+ "codec_model.encoder.layers.4.block.1.conv.weight": "model-00002-of-00002.safetensors",
150
+ "codec_model.encoder.layers.4.block.3.conv.bias": "model-00002-of-00002.safetensors",
151
+ "codec_model.encoder.layers.4.block.3.conv.weight": "model-00002-of-00002.safetensors",
152
+ "codec_model.encoder.layers.6.conv.bias": "model-00002-of-00002.safetensors",
153
+ "codec_model.encoder.layers.6.conv.weight": "model-00002-of-00002.safetensors",
154
+ "codec_model.encoder.layers.7.block.1.conv.bias": "model-00002-of-00002.safetensors",
155
+ "codec_model.encoder.layers.7.block.1.conv.weight": "model-00002-of-00002.safetensors",
156
+ "codec_model.encoder.layers.7.block.3.conv.bias": "model-00002-of-00002.safetensors",
157
+ "codec_model.encoder.layers.7.block.3.conv.weight": "model-00002-of-00002.safetensors",
158
+ "codec_model.encoder.layers.9.conv.bias": "model-00002-of-00002.safetensors",
159
+ "codec_model.encoder.layers.9.conv.weight": "model-00002-of-00002.safetensors",
160
+ "codec_model.encoder_transformer.layers.0.input_layernorm.bias": "model-00002-of-00002.safetensors",
161
+ "codec_model.encoder_transformer.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
162
+ "codec_model.encoder_transformer.layers.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
163
+ "codec_model.encoder_transformer.layers.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
164
+ "codec_model.encoder_transformer.layers.0.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
165
+ "codec_model.encoder_transformer.layers.0.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
166
+ "codec_model.encoder_transformer.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
167
+ "codec_model.encoder_transformer.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
168
+ "codec_model.encoder_transformer.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
169
+ "codec_model.encoder_transformer.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
170
+ "codec_model.encoder_transformer.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
171
+ "codec_model.encoder_transformer.layers.0.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
172
+ "codec_model.encoder_transformer.layers.1.input_layernorm.bias": "model-00002-of-00002.safetensors",
173
+ "codec_model.encoder_transformer.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
174
+ "codec_model.encoder_transformer.layers.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
175
+ "codec_model.encoder_transformer.layers.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
176
+ "codec_model.encoder_transformer.layers.1.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
177
+ "codec_model.encoder_transformer.layers.1.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
178
+ "codec_model.encoder_transformer.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
179
+ "codec_model.encoder_transformer.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
180
+ "codec_model.encoder_transformer.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
181
+ "codec_model.encoder_transformer.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
182
+ "codec_model.encoder_transformer.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
183
+ "codec_model.encoder_transformer.layers.1.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
184
+ "codec_model.encoder_transformer.layers.2.input_layernorm.bias": "model-00002-of-00002.safetensors",
185
+ "codec_model.encoder_transformer.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
186
+ "codec_model.encoder_transformer.layers.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
187
+ "codec_model.encoder_transformer.layers.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
188
+ "codec_model.encoder_transformer.layers.2.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
189
+ "codec_model.encoder_transformer.layers.2.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
190
+ "codec_model.encoder_transformer.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
191
+ "codec_model.encoder_transformer.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
192
+ "codec_model.encoder_transformer.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
193
+ "codec_model.encoder_transformer.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
194
+ "codec_model.encoder_transformer.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
195
+ "codec_model.encoder_transformer.layers.2.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
196
+ "codec_model.encoder_transformer.layers.3.input_layernorm.bias": "model-00002-of-00002.safetensors",
197
+ "codec_model.encoder_transformer.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
198
+ "codec_model.encoder_transformer.layers.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
199
+ "codec_model.encoder_transformer.layers.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
200
+ "codec_model.encoder_transformer.layers.3.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
201
+ "codec_model.encoder_transformer.layers.3.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
202
+ "codec_model.encoder_transformer.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
203
+ "codec_model.encoder_transformer.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
204
+ "codec_model.encoder_transformer.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
205
+ "codec_model.encoder_transformer.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
206
+ "codec_model.encoder_transformer.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
207
+ "codec_model.encoder_transformer.layers.3.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
208
+ "codec_model.encoder_transformer.layers.4.input_layernorm.bias": "model-00002-of-00002.safetensors",
209
+ "codec_model.encoder_transformer.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors",
210
+ "codec_model.encoder_transformer.layers.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
211
+ "codec_model.encoder_transformer.layers.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
212
+ "codec_model.encoder_transformer.layers.4.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
213
+ "codec_model.encoder_transformer.layers.4.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
214
+ "codec_model.encoder_transformer.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
215
+ "codec_model.encoder_transformer.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
216
+ "codec_model.encoder_transformer.layers.4.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
217
+ "codec_model.encoder_transformer.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
218
+ "codec_model.encoder_transformer.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
219
+ "codec_model.encoder_transformer.layers.4.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
220
+ "codec_model.encoder_transformer.layers.5.input_layernorm.bias": "model-00002-of-00002.safetensors",
221
+ "codec_model.encoder_transformer.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors",
222
+ "codec_model.encoder_transformer.layers.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
223
+ "codec_model.encoder_transformer.layers.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
224
+ "codec_model.encoder_transformer.layers.5.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
225
+ "codec_model.encoder_transformer.layers.5.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
226
+ "codec_model.encoder_transformer.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
227
+ "codec_model.encoder_transformer.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
228
+ "codec_model.encoder_transformer.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
229
+ "codec_model.encoder_transformer.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
230
+ "codec_model.encoder_transformer.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
231
+ "codec_model.encoder_transformer.layers.5.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
232
+ "codec_model.encoder_transformer.layers.6.input_layernorm.bias": "model-00002-of-00002.safetensors",
233
+ "codec_model.encoder_transformer.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors",
234
+ "codec_model.encoder_transformer.layers.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
235
+ "codec_model.encoder_transformer.layers.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
236
+ "codec_model.encoder_transformer.layers.6.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
237
+ "codec_model.encoder_transformer.layers.6.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
238
+ "codec_model.encoder_transformer.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
239
+ "codec_model.encoder_transformer.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
240
+ "codec_model.encoder_transformer.layers.6.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
241
+ "codec_model.encoder_transformer.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
242
+ "codec_model.encoder_transformer.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
243
+ "codec_model.encoder_transformer.layers.6.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
244
+ "codec_model.encoder_transformer.layers.7.input_layernorm.bias": "model-00002-of-00002.safetensors",
245
+ "codec_model.encoder_transformer.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors",
246
+ "codec_model.encoder_transformer.layers.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
247
+ "codec_model.encoder_transformer.layers.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
248
+ "codec_model.encoder_transformer.layers.7.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
249
+ "codec_model.encoder_transformer.layers.7.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
250
+ "codec_model.encoder_transformer.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
251
+ "codec_model.encoder_transformer.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
252
+ "codec_model.encoder_transformer.layers.7.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
253
+ "codec_model.encoder_transformer.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
254
+ "codec_model.encoder_transformer.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
255
+ "codec_model.encoder_transformer.layers.7.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
256
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.input_proj.weight": "model-00002-of-00002.safetensors",
257
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.cluster_usage": "model-00002-of-00002.safetensors",
258
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.embed_sum": "model-00002-of-00002.safetensors",
259
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.initialized": "model-00002-of-00002.safetensors",
260
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.cluster_usage": "model-00002-of-00002.safetensors",
261
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.embed_sum": "model-00002-of-00002.safetensors",
262
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.initialized": "model-00002-of-00002.safetensors",
263
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.cluster_usage": "model-00002-of-00002.safetensors",
264
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.embed_sum": "model-00002-of-00002.safetensors",
265
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.initialized": "model-00002-of-00002.safetensors",
266
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.cluster_usage": "model-00002-of-00002.safetensors",
267
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.embed_sum": "model-00002-of-00002.safetensors",
268
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.initialized": "model-00002-of-00002.safetensors",
269
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.cluster_usage": "model-00002-of-00002.safetensors",
270
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.embed_sum": "model-00002-of-00002.safetensors",
271
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.initialized": "model-00002-of-00002.safetensors",
272
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.cluster_usage": "model-00002-of-00002.safetensors",
273
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.embed_sum": "model-00002-of-00002.safetensors",
274
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.initialized": "model-00002-of-00002.safetensors",
275
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.cluster_usage": "model-00002-of-00002.safetensors",
276
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.embed_sum": "model-00002-of-00002.safetensors",
277
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.initialized": "model-00002-of-00002.safetensors",
278
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.cluster_usage": "model-00002-of-00002.safetensors",
279
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.embed_sum": "model-00002-of-00002.safetensors",
280
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.initialized": "model-00002-of-00002.safetensors",
281
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.cluster_usage": "model-00002-of-00002.safetensors",
282
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.embed_sum": "model-00002-of-00002.safetensors",
283
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.initialized": "model-00002-of-00002.safetensors",
284
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.cluster_usage": "model-00002-of-00002.safetensors",
285
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.embed_sum": "model-00002-of-00002.safetensors",
286
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.initialized": "model-00002-of-00002.safetensors",
287
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.cluster_usage": "model-00002-of-00002.safetensors",
288
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.embed_sum": "model-00002-of-00002.safetensors",
289
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.initialized": "model-00002-of-00002.safetensors",
290
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.cluster_usage": "model-00002-of-00002.safetensors",
291
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.embed_sum": "model-00002-of-00002.safetensors",
292
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.initialized": "model-00002-of-00002.safetensors",
293
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.cluster_usage": "model-00002-of-00002.safetensors",
294
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.embed_sum": "model-00002-of-00002.safetensors",
295
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.initialized": "model-00002-of-00002.safetensors",
296
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.cluster_usage": "model-00002-of-00002.safetensors",
297
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.embed_sum": "model-00002-of-00002.safetensors",
298
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.initialized": "model-00002-of-00002.safetensors",
299
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.cluster_usage": "model-00002-of-00002.safetensors",
300
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.embed_sum": "model-00002-of-00002.safetensors",
301
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.initialized": "model-00002-of-00002.safetensors",
302
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.cluster_usage": "model-00002-of-00002.safetensors",
303
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.embed_sum": "model-00002-of-00002.safetensors",
304
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.initialized": "model-00002-of-00002.safetensors",
305
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.cluster_usage": "model-00002-of-00002.safetensors",
306
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.embed_sum": "model-00002-of-00002.safetensors",
307
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.initialized": "model-00002-of-00002.safetensors",
308
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.cluster_usage": "model-00002-of-00002.safetensors",
309
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.embed_sum": "model-00002-of-00002.safetensors",
310
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.initialized": "model-00002-of-00002.safetensors",
311
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.cluster_usage": "model-00002-of-00002.safetensors",
312
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.embed_sum": "model-00002-of-00002.safetensors",
313
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.initialized": "model-00002-of-00002.safetensors",
314
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.cluster_usage": "model-00002-of-00002.safetensors",
315
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.embed_sum": "model-00002-of-00002.safetensors",
316
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.initialized": "model-00002-of-00002.safetensors",
317
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.cluster_usage": "model-00002-of-00002.safetensors",
318
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.embed_sum": "model-00002-of-00002.safetensors",
319
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.initialized": "model-00002-of-00002.safetensors",
320
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.cluster_usage": "model-00002-of-00002.safetensors",
321
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.embed_sum": "model-00002-of-00002.safetensors",
322
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.initialized": "model-00002-of-00002.safetensors",
323
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.cluster_usage": "model-00002-of-00002.safetensors",
324
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.embed_sum": "model-00002-of-00002.safetensors",
325
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.initialized": "model-00002-of-00002.safetensors",
326
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.cluster_usage": "model-00002-of-00002.safetensors",
327
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.embed_sum": "model-00002-of-00002.safetensors",
328
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.initialized": "model-00002-of-00002.safetensors",
329
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.cluster_usage": "model-00002-of-00002.safetensors",
330
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.embed_sum": "model-00002-of-00002.safetensors",
331
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.initialized": "model-00002-of-00002.safetensors",
332
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.cluster_usage": "model-00002-of-00002.safetensors",
333
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.embed_sum": "model-00002-of-00002.safetensors",
334
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.initialized": "model-00002-of-00002.safetensors",
335
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.cluster_usage": "model-00002-of-00002.safetensors",
336
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.embed_sum": "model-00002-of-00002.safetensors",
337
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.initialized": "model-00002-of-00002.safetensors",
338
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.cluster_usage": "model-00002-of-00002.safetensors",
339
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.embed_sum": "model-00002-of-00002.safetensors",
340
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.initialized": "model-00002-of-00002.safetensors",
341
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.cluster_usage": "model-00002-of-00002.safetensors",
342
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.embed_sum": "model-00002-of-00002.safetensors",
343
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.initialized": "model-00002-of-00002.safetensors",
344
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.cluster_usage": "model-00002-of-00002.safetensors",
345
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.embed_sum": "model-00002-of-00002.safetensors",
346
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.initialized": "model-00002-of-00002.safetensors",
347
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.cluster_usage": "model-00002-of-00002.safetensors",
348
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.embed_sum": "model-00002-of-00002.safetensors",
349
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.initialized": "model-00002-of-00002.safetensors",
350
+ "codec_model.quantizer.acoustic_residual_vector_quantizer.output_proj.weight": "model-00002-of-00002.safetensors",
351
+ "codec_model.quantizer.semantic_residual_vector_quantizer.input_proj.weight": "model-00002-of-00002.safetensors",
352
+ "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.cluster_usage": "model-00002-of-00002.safetensors",
353
+ "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.embed_sum": "model-00002-of-00002.safetensors",
354
+ "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.initialized": "model-00002-of-00002.safetensors",
355
+ "codec_model.quantizer.semantic_residual_vector_quantizer.output_proj.weight": "model-00002-of-00002.safetensors",
356
+ "codec_model.upsample.conv.weight": "model-00002-of-00002.safetensors",
357
+ "lm_head.weight": "model-00002-of-00002.safetensors",
358
+ "model.embed_tokens.embed_tokens.weight": "model-00001-of-00002.safetensors",
359
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
360
+ "model.layers.0.mlp.fc1.weight": "model-00001-of-00002.safetensors",
361
+ "model.layers.0.mlp.fc2.weight": "model-00001-of-00002.safetensors",
362
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
363
+ "model.layers.0.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
364
+ "model.layers.0.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
365
+ "model.layers.0.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
366
+ "model.layers.0.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
367
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
368
+ "model.layers.1.mlp.fc1.weight": "model-00001-of-00002.safetensors",
369
+ "model.layers.1.mlp.fc2.weight": "model-00001-of-00002.safetensors",
370
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
371
+ "model.layers.1.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
372
+ "model.layers.1.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
373
+ "model.layers.1.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
374
+ "model.layers.1.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
375
+ "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
376
+ "model.layers.10.mlp.fc1.weight": "model-00001-of-00002.safetensors",
377
+ "model.layers.10.mlp.fc2.weight": "model-00001-of-00002.safetensors",
378
+ "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
379
+ "model.layers.10.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
380
+ "model.layers.10.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
381
+ "model.layers.10.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
382
+ "model.layers.10.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
383
+ "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
384
+ "model.layers.11.mlp.fc1.weight": "model-00001-of-00002.safetensors",
385
+ "model.layers.11.mlp.fc2.weight": "model-00001-of-00002.safetensors",
386
+ "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
387
+ "model.layers.11.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
388
+ "model.layers.11.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
389
+ "model.layers.11.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
390
+ "model.layers.11.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
391
+ "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
392
+ "model.layers.12.mlp.fc1.weight": "model-00001-of-00002.safetensors",
393
+ "model.layers.12.mlp.fc2.weight": "model-00001-of-00002.safetensors",
394
+ "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
395
+ "model.layers.12.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
396
+ "model.layers.12.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
397
+ "model.layers.12.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
398
+ "model.layers.12.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
399
+ "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
400
+ "model.layers.13.mlp.fc1.weight": "model-00001-of-00002.safetensors",
401
+ "model.layers.13.mlp.fc2.weight": "model-00001-of-00002.safetensors",
402
+ "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
403
+ "model.layers.13.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
404
+ "model.layers.13.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
405
+ "model.layers.13.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
406
+ "model.layers.13.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
407
+ "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
408
+ "model.layers.14.mlp.fc1.weight": "model-00001-of-00002.safetensors",
409
+ "model.layers.14.mlp.fc2.weight": "model-00001-of-00002.safetensors",
410
+ "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
411
+ "model.layers.14.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
412
+ "model.layers.14.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
413
+ "model.layers.14.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
414
+ "model.layers.14.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
415
+ "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
416
+ "model.layers.15.mlp.fc1.weight": "model-00001-of-00002.safetensors",
417
+ "model.layers.15.mlp.fc2.weight": "model-00001-of-00002.safetensors",
418
+ "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
419
+ "model.layers.15.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
420
+ "model.layers.15.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
421
+ "model.layers.15.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
422
+ "model.layers.15.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
423
+ "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
424
+ "model.layers.16.mlp.fc1.weight": "model-00001-of-00002.safetensors",
425
+ "model.layers.16.mlp.fc2.weight": "model-00001-of-00002.safetensors",
426
+ "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
427
+ "model.layers.16.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
428
+ "model.layers.16.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
429
+ "model.layers.16.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
430
+ "model.layers.16.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
431
+ "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
432
+ "model.layers.17.mlp.fc1.weight": "model-00001-of-00002.safetensors",
433
+ "model.layers.17.mlp.fc2.weight": "model-00001-of-00002.safetensors",
434
+ "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
435
+ "model.layers.17.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
436
+ "model.layers.17.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
437
+ "model.layers.17.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
438
+ "model.layers.17.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
439
+ "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
440
+ "model.layers.18.mlp.fc1.weight": "model-00001-of-00002.safetensors",
441
+ "model.layers.18.mlp.fc2.weight": "model-00001-of-00002.safetensors",
442
+ "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
443
+ "model.layers.18.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
444
+ "model.layers.18.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
445
+ "model.layers.18.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
446
+ "model.layers.18.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
447
+ "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
448
+ "model.layers.19.mlp.fc1.weight": "model-00001-of-00002.safetensors",
449
+ "model.layers.19.mlp.fc2.weight": "model-00001-of-00002.safetensors",
450
+ "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
451
+ "model.layers.19.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
452
+ "model.layers.19.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
453
+ "model.layers.19.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
454
+ "model.layers.19.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
455
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
456
+ "model.layers.2.mlp.fc1.weight": "model-00001-of-00002.safetensors",
457
+ "model.layers.2.mlp.fc2.weight": "model-00001-of-00002.safetensors",
458
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
459
+ "model.layers.2.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
460
+ "model.layers.2.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
461
+ "model.layers.2.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
462
+ "model.layers.2.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
463
+ "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
464
+ "model.layers.20.mlp.fc1.weight": "model-00001-of-00002.safetensors",
465
+ "model.layers.20.mlp.fc2.weight": "model-00001-of-00002.safetensors",
466
+ "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
467
+ "model.layers.20.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
468
+ "model.layers.20.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
469
+ "model.layers.20.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
470
+ "model.layers.20.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
471
+ "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
472
+ "model.layers.21.mlp.fc1.weight": "model-00001-of-00002.safetensors",
473
+ "model.layers.21.mlp.fc2.weight": "model-00001-of-00002.safetensors",
474
+ "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
475
+ "model.layers.21.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
476
+ "model.layers.21.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
477
+ "model.layers.21.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
478
+ "model.layers.21.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
479
+ "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
480
+ "model.layers.22.mlp.fc1.weight": "model-00001-of-00002.safetensors",
481
+ "model.layers.22.mlp.fc2.weight": "model-00001-of-00002.safetensors",
482
+ "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
483
+ "model.layers.22.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
484
+ "model.layers.22.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
485
+ "model.layers.22.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
486
+ "model.layers.22.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
487
+ "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
488
+ "model.layers.23.mlp.fc1.weight": "model-00001-of-00002.safetensors",
489
+ "model.layers.23.mlp.fc2.weight": "model-00001-of-00002.safetensors",
490
+ "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
491
+ "model.layers.23.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
492
+ "model.layers.23.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
493
+ "model.layers.23.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
494
+ "model.layers.23.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
495
+ "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
496
+ "model.layers.24.mlp.fc1.weight": "model-00001-of-00002.safetensors",
497
+ "model.layers.24.mlp.fc2.weight": "model-00001-of-00002.safetensors",
498
+ "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
499
+ "model.layers.24.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
500
+ "model.layers.24.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
501
+ "model.layers.24.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
502
+ "model.layers.24.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
503
+ "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
504
+ "model.layers.25.mlp.fc1.weight": "model-00001-of-00002.safetensors",
505
+ "model.layers.25.mlp.fc2.weight": "model-00001-of-00002.safetensors",
506
+ "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
507
+ "model.layers.25.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
508
+ "model.layers.25.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
509
+ "model.layers.25.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
510
+ "model.layers.25.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
511
+ "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
512
+ "model.layers.26.mlp.fc1.weight": "model-00001-of-00002.safetensors",
513
+ "model.layers.26.mlp.fc2.weight": "model-00001-of-00002.safetensors",
514
+ "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
515
+ "model.layers.26.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
516
+ "model.layers.26.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
517
+ "model.layers.26.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
518
+ "model.layers.26.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
519
+ "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
520
+ "model.layers.27.mlp.fc1.weight": "model-00001-of-00002.safetensors",
521
+ "model.layers.27.mlp.fc2.weight": "model-00001-of-00002.safetensors",
522
+ "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
523
+ "model.layers.27.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
524
+ "model.layers.27.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
525
+ "model.layers.27.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
526
+ "model.layers.27.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
527
+ "model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors",
528
+ "model.layers.28.mlp.fc1.weight": "model-00001-of-00002.safetensors",
529
+ "model.layers.28.mlp.fc2.weight": "model-00001-of-00002.safetensors",
530
+ "model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
531
+ "model.layers.28.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
532
+ "model.layers.28.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
533
+ "model.layers.28.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
534
+ "model.layers.28.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
535
+ "model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors",
536
+ "model.layers.29.mlp.fc1.weight": "model-00001-of-00002.safetensors",
537
+ "model.layers.29.mlp.fc2.weight": "model-00001-of-00002.safetensors",
538
+ "model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
539
+ "model.layers.29.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
540
+ "model.layers.29.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
541
+ "model.layers.29.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
542
+ "model.layers.29.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
543
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
544
+ "model.layers.3.mlp.fc1.weight": "model-00001-of-00002.safetensors",
545
+ "model.layers.3.mlp.fc2.weight": "model-00001-of-00002.safetensors",
546
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
547
+ "model.layers.3.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
548
+ "model.layers.3.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
549
+ "model.layers.3.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
550
+ "model.layers.3.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
551
+ "model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors",
552
+ "model.layers.30.mlp.fc1.weight": "model-00001-of-00002.safetensors",
553
+ "model.layers.30.mlp.fc2.weight": "model-00001-of-00002.safetensors",
554
+ "model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
555
+ "model.layers.30.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
556
+ "model.layers.30.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
557
+ "model.layers.30.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
558
+ "model.layers.30.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
559
+ "model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors",
560
+ "model.layers.31.mlp.fc1.weight": "model-00001-of-00002.safetensors",
561
+ "model.layers.31.mlp.fc2.weight": "model-00001-of-00002.safetensors",
562
+ "model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
563
+ "model.layers.31.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
564
+ "model.layers.31.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
565
+ "model.layers.31.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
566
+ "model.layers.31.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
567
+ "model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors",
568
+ "model.layers.32.mlp.fc1.weight": "model-00001-of-00002.safetensors",
569
+ "model.layers.32.mlp.fc2.weight": "model-00001-of-00002.safetensors",
570
+ "model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
571
+ "model.layers.32.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
572
+ "model.layers.32.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
573
+ "model.layers.32.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
574
+ "model.layers.32.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
575
+ "model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors",
576
+ "model.layers.33.mlp.fc1.weight": "model-00001-of-00002.safetensors",
577
+ "model.layers.33.mlp.fc2.weight": "model-00001-of-00002.safetensors",
578
+ "model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
579
+ "model.layers.33.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
580
+ "model.layers.33.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
581
+ "model.layers.33.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
582
+ "model.layers.33.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
583
+ "model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors",
584
+ "model.layers.34.mlp.fc1.weight": "model-00001-of-00002.safetensors",
585
+ "model.layers.34.mlp.fc2.weight": "model-00001-of-00002.safetensors",
586
+ "model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
587
+ "model.layers.34.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
588
+ "model.layers.34.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
589
+ "model.layers.34.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
590
+ "model.layers.34.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
591
+ "model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors",
592
+ "model.layers.35.mlp.fc1.weight": "model-00001-of-00002.safetensors",
593
+ "model.layers.35.mlp.fc2.weight": "model-00001-of-00002.safetensors",
594
+ "model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
595
+ "model.layers.35.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
596
+ "model.layers.35.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
597
+ "model.layers.35.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
598
+ "model.layers.35.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
599
+ "model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors",
600
+ "model.layers.36.mlp.fc1.weight": "model-00001-of-00002.safetensors",
601
+ "model.layers.36.mlp.fc2.weight": "model-00001-of-00002.safetensors",
602
+ "model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
603
+ "model.layers.36.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
604
+ "model.layers.36.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
605
+ "model.layers.36.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
606
+ "model.layers.36.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
607
+ "model.layers.37.input_layernorm.weight": "model-00002-of-00002.safetensors",
608
+ "model.layers.37.mlp.fc1.weight": "model-00002-of-00002.safetensors",
609
+ "model.layers.37.mlp.fc2.weight": "model-00002-of-00002.safetensors",
610
+ "model.layers.37.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
611
+ "model.layers.37.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
612
+ "model.layers.37.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
613
+ "model.layers.37.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
614
+ "model.layers.37.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
615
+ "model.layers.38.input_layernorm.weight": "model-00002-of-00002.safetensors",
616
+ "model.layers.38.mlp.fc1.weight": "model-00002-of-00002.safetensors",
617
+ "model.layers.38.mlp.fc2.weight": "model-00002-of-00002.safetensors",
618
+ "model.layers.38.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
619
+ "model.layers.38.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
620
+ "model.layers.38.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
621
+ "model.layers.38.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
622
+ "model.layers.38.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
623
+ "model.layers.39.input_layernorm.weight": "model-00002-of-00002.safetensors",
624
+ "model.layers.39.mlp.fc1.weight": "model-00002-of-00002.safetensors",
625
+ "model.layers.39.mlp.fc2.weight": "model-00002-of-00002.safetensors",
626
+ "model.layers.39.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
627
+ "model.layers.39.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
628
+ "model.layers.39.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
629
+ "model.layers.39.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
630
+ "model.layers.39.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
631
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
632
+ "model.layers.4.mlp.fc1.weight": "model-00001-of-00002.safetensors",
633
+ "model.layers.4.mlp.fc2.weight": "model-00001-of-00002.safetensors",
634
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
635
+ "model.layers.4.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
636
+ "model.layers.4.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
637
+ "model.layers.4.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
638
+ "model.layers.4.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
639
+ "model.layers.40.input_layernorm.weight": "model-00002-of-00002.safetensors",
640
+ "model.layers.40.mlp.fc1.weight": "model-00002-of-00002.safetensors",
641
+ "model.layers.40.mlp.fc2.weight": "model-00002-of-00002.safetensors",
642
+ "model.layers.40.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
643
+ "model.layers.40.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
644
+ "model.layers.40.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
645
+ "model.layers.40.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
646
+ "model.layers.40.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
647
+ "model.layers.41.input_layernorm.weight": "model-00002-of-00002.safetensors",
648
+ "model.layers.41.mlp.fc1.weight": "model-00002-of-00002.safetensors",
649
+ "model.layers.41.mlp.fc2.weight": "model-00002-of-00002.safetensors",
650
+ "model.layers.41.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
651
+ "model.layers.41.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
652
+ "model.layers.41.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
653
+ "model.layers.41.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
654
+ "model.layers.41.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
655
+ "model.layers.42.input_layernorm.weight": "model-00002-of-00002.safetensors",
656
+ "model.layers.42.mlp.fc1.weight": "model-00002-of-00002.safetensors",
657
+ "model.layers.42.mlp.fc2.weight": "model-00002-of-00002.safetensors",
658
+ "model.layers.42.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
659
+ "model.layers.42.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
660
+ "model.layers.42.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
661
+ "model.layers.42.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
662
+ "model.layers.42.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
663
+ "model.layers.43.input_layernorm.weight": "model-00002-of-00002.safetensors",
664
+ "model.layers.43.mlp.fc1.weight": "model-00002-of-00002.safetensors",
665
+ "model.layers.43.mlp.fc2.weight": "model-00002-of-00002.safetensors",
666
+ "model.layers.43.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
667
+ "model.layers.43.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
668
+ "model.layers.43.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
669
+ "model.layers.43.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
670
+ "model.layers.43.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
671
+ "model.layers.44.input_layernorm.weight": "model-00002-of-00002.safetensors",
672
+ "model.layers.44.mlp.fc1.weight": "model-00002-of-00002.safetensors",
673
+ "model.layers.44.mlp.fc2.weight": "model-00002-of-00002.safetensors",
674
+ "model.layers.44.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
675
+ "model.layers.44.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
676
+ "model.layers.44.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
677
+ "model.layers.44.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
678
+ "model.layers.44.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
679
+ "model.layers.45.input_layernorm.weight": "model-00002-of-00002.safetensors",
680
+ "model.layers.45.mlp.fc1.weight": "model-00002-of-00002.safetensors",
681
+ "model.layers.45.mlp.fc2.weight": "model-00002-of-00002.safetensors",
682
+ "model.layers.45.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
683
+ "model.layers.45.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
684
+ "model.layers.45.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
685
+ "model.layers.45.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
686
+ "model.layers.45.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
687
+ "model.layers.46.input_layernorm.weight": "model-00002-of-00002.safetensors",
688
+ "model.layers.46.mlp.fc1.weight": "model-00002-of-00002.safetensors",
689
+ "model.layers.46.mlp.fc2.weight": "model-00002-of-00002.safetensors",
690
+ "model.layers.46.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
691
+ "model.layers.46.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
692
+ "model.layers.46.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
693
+ "model.layers.46.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
694
+ "model.layers.46.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
695
+ "model.layers.47.input_layernorm.weight": "model-00002-of-00002.safetensors",
696
+ "model.layers.47.mlp.fc1.weight": "model-00002-of-00002.safetensors",
697
+ "model.layers.47.mlp.fc2.weight": "model-00002-of-00002.safetensors",
698
+ "model.layers.47.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
699
+ "model.layers.47.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors",
700
+ "model.layers.47.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors",
701
+ "model.layers.47.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors",
702
+ "model.layers.47.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
703
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
704
+ "model.layers.5.mlp.fc1.weight": "model-00001-of-00002.safetensors",
705
+ "model.layers.5.mlp.fc2.weight": "model-00001-of-00002.safetensors",
706
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
707
+ "model.layers.5.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
708
+ "model.layers.5.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
709
+ "model.layers.5.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
710
+ "model.layers.5.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
711
+ "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
712
+ "model.layers.6.mlp.fc1.weight": "model-00001-of-00002.safetensors",
713
+ "model.layers.6.mlp.fc2.weight": "model-00001-of-00002.safetensors",
714
+ "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
715
+ "model.layers.6.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
716
+ "model.layers.6.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
717
+ "model.layers.6.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
718
+ "model.layers.6.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
719
+ "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
720
+ "model.layers.7.mlp.fc1.weight": "model-00001-of-00002.safetensors",
721
+ "model.layers.7.mlp.fc2.weight": "model-00001-of-00002.safetensors",
722
+ "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
723
+ "model.layers.7.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
724
+ "model.layers.7.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
725
+ "model.layers.7.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
726
+ "model.layers.7.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
727
+ "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
728
+ "model.layers.8.mlp.fc1.weight": "model-00001-of-00002.safetensors",
729
+ "model.layers.8.mlp.fc2.weight": "model-00001-of-00002.safetensors",
730
+ "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
731
+ "model.layers.8.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
732
+ "model.layers.8.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
733
+ "model.layers.8.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
734
+ "model.layers.8.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
735
+ "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
736
+ "model.layers.9.mlp.fc1.weight": "model-00001-of-00002.safetensors",
737
+ "model.layers.9.mlp.fc2.weight": "model-00001-of-00002.safetensors",
738
+ "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
739
+ "model.layers.9.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
740
+ "model.layers.9.self_attn.o_proj.linear.weight": "model-00001-of-00002.safetensors",
741
+ "model.layers.9.self_attn.q_proj.linear.weight": "model-00001-of-00002.safetensors",
742
+ "model.layers.9.self_attn.v_proj.linear.weight": "model-00001-of-00002.safetensors",
743
+ "model.norm.weight": "model-00002-of-00002.safetensors"
744
+ }
745
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_delay_seconds": 2.5,
3
+ "audio_silence_prefix_seconds": 1.0,
4
+ "chunk_length_s": null,
5
+ "feature_extractor_type": "KyutaiSpeechToTextFeatureExtractor",
6
+ "feature_size": 1,
7
+ "overlap": null,
8
+ "padding_side": "right",
9
+ "padding_value": 0.0,
10
+ "processor_class": "KyutaiSpeechToTextProcessor",
11
+ "return_attention_mask": true,
12
+ "sampling_rate": 24000
13
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "unk_token": "<unk>"
3
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<pad>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token_id": null,
37
+ "chat_template": null,
38
+ "clean_up_tokenization_spaces": false,
39
+ "eos_token_id": null,
40
+ "extra_special_tokens": {},
41
+ "model_input_names": [
42
+ "input_ids",
43
+ "attention_mask"
44
+ ],
45
+ "model_max_length": 1000000000000000019884624838656,
46
+ "pad_token_id": null,
47
+ "processor_class": "KyutaiSpeechToTextProcessor",
48
+ "tokenizer_class": "PreTrainedTokenizerFast",
49
+ "unk_token": "<unk>"
50
+ }