Checkpoint checkpoint-epoch-15
Browse files
checkpoint-epoch-15/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen3-TTS-12Hz-1.7B-Base
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- qwen3-tts
|
| 6 |
+
- lora
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
LoRA adapter checkpoint for epoch 15.
|
checkpoint-epoch-15/adapter_config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/workspace/.hf_home/hub/models--Qwen--Qwen3-TTS-12Hz-1.7B-Base/snapshots/fd4b254389122332181a7c3db7f27e918eec64e3",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": true,
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.05,
|
| 22 |
+
"megatron_config": null,
|
| 23 |
+
"megatron_core": "megatron.core",
|
| 24 |
+
"modules_to_save": null,
|
| 25 |
+
"peft_type": "LORA",
|
| 26 |
+
"peft_version": "0.18.1",
|
| 27 |
+
"qalora_group_size": 16,
|
| 28 |
+
"r": 16,
|
| 29 |
+
"rank_pattern": {},
|
| 30 |
+
"revision": null,
|
| 31 |
+
"target_modules": [
|
| 32 |
+
"gate_proj",
|
| 33 |
+
"v_proj",
|
| 34 |
+
"o_proj",
|
| 35 |
+
"down_proj",
|
| 36 |
+
"up_proj",
|
| 37 |
+
"q_proj",
|
| 38 |
+
"k_proj"
|
| 39 |
+
],
|
| 40 |
+
"target_parameters": null,
|
| 41 |
+
"task_type": "CAUSAL_LM",
|
| 42 |
+
"trainable_token_indices": null,
|
| 43 |
+
"use_dora": false,
|
| 44 |
+
"use_qalora": false,
|
| 45 |
+
"use_rslora": false
|
| 46 |
+
}
|
checkpoint-epoch-15/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:342536bbfbb77ed07a56f6052410a3b2133cfc33dd46e0edb985c40020761fc3
|
| 3 |
+
size 77004832
|
checkpoint-epoch-15/config.json
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3TTSForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"assistant_token_id": 77091,
|
| 6 |
+
"im_end_token_id": 151645,
|
| 7 |
+
"im_start_token_id": 151644,
|
| 8 |
+
"tts_bos_token_id": 151672,
|
| 9 |
+
"tts_eos_token_id": 151673,
|
| 10 |
+
"tts_pad_token_id": 151671,
|
| 11 |
+
"model_type": "qwen3_tts",
|
| 12 |
+
"tokenizer_type": "qwen3_tts_tokenizer_12hz",
|
| 13 |
+
"tts_model_size": "1b7",
|
| 14 |
+
"tts_model_type": "custom_voice",
|
| 15 |
+
"speaker_encoder_config": {
|
| 16 |
+
"enc_dim": 2048,
|
| 17 |
+
"sample_rate": 24000
|
| 18 |
+
},
|
| 19 |
+
"talker_config": {
|
| 20 |
+
"attention_bias": false,
|
| 21 |
+
"attention_dropout": 0,
|
| 22 |
+
"code_predictor_config": {
|
| 23 |
+
"_name_or_path": "",
|
| 24 |
+
"add_cross_attention": false,
|
| 25 |
+
"architectures": null,
|
| 26 |
+
"attention_bias": false,
|
| 27 |
+
"attention_dropout": 0,
|
| 28 |
+
"bad_words_ids": null,
|
| 29 |
+
"begin_suppress_tokens": null,
|
| 30 |
+
"bos_token_id": null,
|
| 31 |
+
"chunk_size_feed_forward": 0,
|
| 32 |
+
"cross_attention_hidden_size": null,
|
| 33 |
+
"decoder_start_token_id": null,
|
| 34 |
+
"diversity_penalty": 0.0,
|
| 35 |
+
"do_sample": false,
|
| 36 |
+
"early_stopping": false,
|
| 37 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 38 |
+
"eos_token_id": null,
|
| 39 |
+
"exponential_decay_length_penalty": null,
|
| 40 |
+
"finetuning_task": null,
|
| 41 |
+
"forced_bos_token_id": null,
|
| 42 |
+
"forced_eos_token_id": null,
|
| 43 |
+
"head_dim": 128,
|
| 44 |
+
"hidden_act": "silu",
|
| 45 |
+
"hidden_size": 1024,
|
| 46 |
+
"id2label": {
|
| 47 |
+
"0": "LABEL_0",
|
| 48 |
+
"1": "LABEL_1"
|
| 49 |
+
},
|
| 50 |
+
"initializer_range": 0.02,
|
| 51 |
+
"intermediate_size": 3072,
|
| 52 |
+
"is_decoder": false,
|
| 53 |
+
"is_encoder_decoder": false,
|
| 54 |
+
"label2id": {
|
| 55 |
+
"LABEL_0": 0,
|
| 56 |
+
"LABEL_1": 1
|
| 57 |
+
},
|
| 58 |
+
"layer_types": [
|
| 59 |
+
"full_attention",
|
| 60 |
+
"full_attention",
|
| 61 |
+
"full_attention",
|
| 62 |
+
"full_attention",
|
| 63 |
+
"full_attention"
|
| 64 |
+
],
|
| 65 |
+
"length_penalty": 1.0,
|
| 66 |
+
"max_length": 20,
|
| 67 |
+
"max_position_embeddings": 65536,
|
| 68 |
+
"max_window_layers": 28,
|
| 69 |
+
"min_length": 0,
|
| 70 |
+
"model_type": "qwen3_tts_talker_code_predictor",
|
| 71 |
+
"no_repeat_ngram_size": 0,
|
| 72 |
+
"num_attention_heads": 16,
|
| 73 |
+
"num_beam_groups": 1,
|
| 74 |
+
"num_beams": 1,
|
| 75 |
+
"num_code_groups": 16,
|
| 76 |
+
"num_hidden_layers": 5,
|
| 77 |
+
"num_key_value_heads": 8,
|
| 78 |
+
"num_return_sequences": 1,
|
| 79 |
+
"output_attentions": false,
|
| 80 |
+
"output_hidden_states": false,
|
| 81 |
+
"output_scores": false,
|
| 82 |
+
"pad_token_id": null,
|
| 83 |
+
"prefix": null,
|
| 84 |
+
"problem_type": null,
|
| 85 |
+
"pruned_heads": {},
|
| 86 |
+
"remove_invalid_values": false,
|
| 87 |
+
"repetition_penalty": 1.0,
|
| 88 |
+
"return_dict": true,
|
| 89 |
+
"return_dict_in_generate": false,
|
| 90 |
+
"rms_norm_eps": 1e-06,
|
| 91 |
+
"rope_scaling": null,
|
| 92 |
+
"rope_theta": 1000000,
|
| 93 |
+
"sep_token_id": null,
|
| 94 |
+
"sliding_window": null,
|
| 95 |
+
"suppress_tokens": null,
|
| 96 |
+
"task_specific_params": null,
|
| 97 |
+
"temperature": 1.0,
|
| 98 |
+
"tf_legacy_loss": false,
|
| 99 |
+
"tie_encoder_decoder": false,
|
| 100 |
+
"tie_word_embeddings": false,
|
| 101 |
+
"tokenizer_class": null,
|
| 102 |
+
"top_k": 50,
|
| 103 |
+
"top_p": 1.0,
|
| 104 |
+
"dtype": null,
|
| 105 |
+
"torchscript": false,
|
| 106 |
+
"typical_p": 1.0,
|
| 107 |
+
"use_bfloat16": false,
|
| 108 |
+
"use_cache": true,
|
| 109 |
+
"use_sliding_window": false,
|
| 110 |
+
"vocab_size": 2048
|
| 111 |
+
},
|
| 112 |
+
"codec_bos_id": 2149,
|
| 113 |
+
"codec_eos_token_id": 2150,
|
| 114 |
+
"codec_think_id": 2154,
|
| 115 |
+
"codec_language_id": {
|
| 116 |
+
"chinese": 2055,
|
| 117 |
+
"english": 2050,
|
| 118 |
+
"german": 2053,
|
| 119 |
+
"italian": 2070,
|
| 120 |
+
"portuguese": 2071,
|
| 121 |
+
"spanish": 2054,
|
| 122 |
+
"japanese": 2058,
|
| 123 |
+
"korean": 2064,
|
| 124 |
+
"french": 2061,
|
| 125 |
+
"russian": 2069
|
| 126 |
+
},
|
| 127 |
+
"codec_nothink_id": 2155,
|
| 128 |
+
"codec_pad_id": 2148,
|
| 129 |
+
"codec_think_bos_id": 2156,
|
| 130 |
+
"codec_think_eos_id": 2157,
|
| 131 |
+
"spk_id": {
|
| 132 |
+
"speaker": 3000
|
| 133 |
+
},
|
| 134 |
+
"spk_is_dialect": {
|
| 135 |
+
"speaker": false
|
| 136 |
+
},
|
| 137 |
+
"head_dim": 128,
|
| 138 |
+
"hidden_act": "silu",
|
| 139 |
+
"hidden_size": 2048,
|
| 140 |
+
"initializer_range": 0.02,
|
| 141 |
+
"intermediate_size": 6144,
|
| 142 |
+
"max_position_embeddings": 32768,
|
| 143 |
+
"model_type": "qwen3_tts_talker",
|
| 144 |
+
"num_attention_heads": 16,
|
| 145 |
+
"num_code_groups": 16,
|
| 146 |
+
"num_hidden_layers": 28,
|
| 147 |
+
"num_key_value_heads": 8,
|
| 148 |
+
"position_id_per_seconds": 13,
|
| 149 |
+
"rms_norm_eps": 1e-06,
|
| 150 |
+
"rope_scaling": {
|
| 151 |
+
"interleaved": true,
|
| 152 |
+
"mrope_section": [
|
| 153 |
+
24,
|
| 154 |
+
20,
|
| 155 |
+
20
|
| 156 |
+
],
|
| 157 |
+
"rope_type": "default",
|
| 158 |
+
"type": "default"
|
| 159 |
+
},
|
| 160 |
+
"rope_theta": 1000000,
|
| 161 |
+
"sliding_window": null,
|
| 162 |
+
"text_hidden_size": 2048,
|
| 163 |
+
"text_vocab_size": 151936,
|
| 164 |
+
"use_cache": true,
|
| 165 |
+
"use_sliding_window": false,
|
| 166 |
+
"vocab_size": 3072
|
| 167 |
+
},
|
| 168 |
+
"transformers_version": "4.57.3"
|
| 169 |
+
}
|
checkpoint-epoch-15/speaker_embedding.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fecc394200a99ade946242f6571aee8e29a3df7bccb2a5591088a86ca7071033
|
| 3 |
+
size 4192
|