checkpoint step 1000
Browse files- checkpoint-1000/config.json +85 -0
- checkpoint-1000/model.safetensors +3 -0
- checkpoint-1000/optimizer.bin +3 -0
- checkpoint-1000/random_states_0.pkl +3 -0
- checkpoint-1000/scheduler.bin +3 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer_config.json +19 -0
- checkpoint-1000/train_config.json +56 -0
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"OmniVoice"
|
| 4 |
+
],
|
| 5 |
+
"audio_codebook_weights": [
|
| 6 |
+
8,
|
| 7 |
+
8,
|
| 8 |
+
6,
|
| 9 |
+
6,
|
| 10 |
+
4,
|
| 11 |
+
4,
|
| 12 |
+
2,
|
| 13 |
+
2
|
| 14 |
+
],
|
| 15 |
+
"audio_mask_id": 1024,
|
| 16 |
+
"audio_vocab_size": 1025,
|
| 17 |
+
"bos_token_id": 2,
|
| 18 |
+
"dtype": "float32",
|
| 19 |
+
"eos_token_id": 3,
|
| 20 |
+
"llm_config": {
|
| 21 |
+
"_name_or_path": "/content/OmniVoice/abcd/checkpoint-6000",
|
| 22 |
+
"architectures": [
|
| 23 |
+
"Qwen3ForCausalLM"
|
| 24 |
+
],
|
| 25 |
+
"attention_bias": false,
|
| 26 |
+
"attention_dropout": 0.0,
|
| 27 |
+
"bos_token_id": 2,
|
| 28 |
+
"chunk_size_feed_forward": 0,
|
| 29 |
+
"dtype": "float32",
|
| 30 |
+
"eos_token_id": 3,
|
| 31 |
+
"head_dim": 64,
|
| 32 |
+
"hidden_act": "silu",
|
| 33 |
+
"hidden_size": 512,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1"
|
| 37 |
+
},
|
| 38 |
+
"initializer_range": 0.02,
|
| 39 |
+
"intermediate_size": 2048,
|
| 40 |
+
"is_encoder_decoder": false,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1
|
| 44 |
+
},
|
| 45 |
+
"layer_types": [
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"full_attention",
|
| 54 |
+
"full_attention",
|
| 55 |
+
"full_attention",
|
| 56 |
+
"full_attention",
|
| 57 |
+
"full_attention"
|
| 58 |
+
],
|
| 59 |
+
"max_position_embeddings": 2048,
|
| 60 |
+
"max_window_layers": 28,
|
| 61 |
+
"model_type": "qwen3",
|
| 62 |
+
"num_attention_heads": 8,
|
| 63 |
+
"num_hidden_layers": 12,
|
| 64 |
+
"num_key_value_heads": 1,
|
| 65 |
+
"output_attentions": false,
|
| 66 |
+
"output_hidden_states": false,
|
| 67 |
+
"pad_token_id": 0,
|
| 68 |
+
"problem_type": null,
|
| 69 |
+
"return_dict": true,
|
| 70 |
+
"rms_norm_eps": 1e-06,
|
| 71 |
+
"rope_parameters": {
|
| 72 |
+
"rope_theta": 20000,
|
| 73 |
+
"rope_type": "default"
|
| 74 |
+
},
|
| 75 |
+
"sliding_window": null,
|
| 76 |
+
"tie_word_embeddings": true,
|
| 77 |
+
"use_cache": false,
|
| 78 |
+
"use_sliding_window": false,
|
| 79 |
+
"vocab_size": 50007
|
| 80 |
+
},
|
| 81 |
+
"model_type": "omnivoice",
|
| 82 |
+
"num_audio_codebook": 8,
|
| 83 |
+
"pad_token_id": 0,
|
| 84 |
+
"transformers_version": "5.5.4"
|
| 85 |
+
}
|
checkpoint-1000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66757f64a30724d0ed5bc9931478cac0d5008bacf3fc30497b0a5e1ca0cb2855
|
| 3 |
+
size 315380216
|
checkpoint-1000/optimizer.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a91efa96256cfb66a8aa721bdc06b618ca60ea49551ec8ba3d7bf1665a8259cb
|
| 3 |
+
size 630845323
|
checkpoint-1000/random_states_0.pkl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b17a7554e51a0a3f0489003c6fca4b907828d5a539165ac79587b2c44ac1e44
|
| 3 |
+
size 14821
|
checkpoint-1000/scheduler.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d0abf748562e21a1d641333a9ddaeb256e898e20b3d8cc9971b76a5488e7ea8
|
| 3 |
+
size 1465
|
checkpoint-1000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|bos|>",
|
| 4 |
+
"eos_token": "<|eos|>",
|
| 5 |
+
"extra_special_tokens": [
|
| 6 |
+
"<|denoise|>",
|
| 7 |
+
"<|lang_start|>",
|
| 8 |
+
"<|lang_end|>",
|
| 9 |
+
"<|instruct_start|>",
|
| 10 |
+
"<|instruct_end|>",
|
| 11 |
+
"<|text_start|>",
|
| 12 |
+
"<|text_end|>"
|
| 13 |
+
],
|
| 14 |
+
"is_local": true,
|
| 15 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 16 |
+
"pad_token": "<|pad|>",
|
| 17 |
+
"tokenizer_class": "TokenizersBackend",
|
| 18 |
+
"unk_token": "<|unk|>"
|
| 19 |
+
}
|
checkpoint-1000/train_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"output_dir": "exp/omnivoice_urdu",
|
| 3 |
+
"data_config": "examples/config/data_config_urdu_resolved.json",
|
| 4 |
+
"llm_name_or_path": "/content/OmniVoice/abcd/checkpoint-6000",
|
| 5 |
+
"audio_vocab_size": 1025,
|
| 6 |
+
"audio_mask_id": 1024,
|
| 7 |
+
"num_audio_codebook": 8,
|
| 8 |
+
"audio_codebook_weights": [
|
| 9 |
+
8,
|
| 10 |
+
8,
|
| 11 |
+
6,
|
| 12 |
+
6,
|
| 13 |
+
4,
|
| 14 |
+
4,
|
| 15 |
+
2,
|
| 16 |
+
2
|
| 17 |
+
],
|
| 18 |
+
"drop_cond_ratio": 0.1,
|
| 19 |
+
"prompt_ratio_range": [
|
| 20 |
+
0.0,
|
| 21 |
+
0.3
|
| 22 |
+
],
|
| 23 |
+
"mask_ratio_range": [
|
| 24 |
+
0.0,
|
| 25 |
+
1.0
|
| 26 |
+
],
|
| 27 |
+
"language_ratio": 0.0,
|
| 28 |
+
"use_pinyin_ratio": 0.0,
|
| 29 |
+
"instruct_ratio": 0.0,
|
| 30 |
+
"only_instruct_ratio": 0.0,
|
| 31 |
+
"resume_from_checkpoint": null,
|
| 32 |
+
"init_from_checkpoint": null,
|
| 33 |
+
"learning_rate": 0.0001,
|
| 34 |
+
"weight_decay": 0.01,
|
| 35 |
+
"max_grad_norm": 1.0,
|
| 36 |
+
"steps": 300000,
|
| 37 |
+
"seed": 42,
|
| 38 |
+
"lr_scheduler_type": "cosine",
|
| 39 |
+
"warmup_type": "ratio",
|
| 40 |
+
"warmup_ratio": 0.03,
|
| 41 |
+
"warmup_steps": 0,
|
| 42 |
+
"batch_tokens": 8192,
|
| 43 |
+
"gradient_accumulation_steps": 1,
|
| 44 |
+
"num_workers": 2,
|
| 45 |
+
"mixed_precision": "no",
|
| 46 |
+
"allow_tf32": false,
|
| 47 |
+
"use_deepspeed": false,
|
| 48 |
+
"deepspeed_config": null,
|
| 49 |
+
"logging_steps": 50,
|
| 50 |
+
"eval_steps": 500,
|
| 51 |
+
"save_steps": 1000,
|
| 52 |
+
"keep_last_n_checkpoints": -1,
|
| 53 |
+
"push_to_hub": true,
|
| 54 |
+
"hub_model_id": "Humair332/omnivoice-urdu",
|
| 55 |
+
"hub_token": null
|
| 56 |
+
}
|