output_dir string | data_config string | llm_name_or_path string | audio_vocab_size int64 | audio_mask_id int64 | num_audio_codebook int64 | audio_codebook_weights list | drop_cond_ratio float64 | prompt_ratio_range list | mask_ratio_range list | language_ratio float64 | use_pinyin_ratio float64 | instruct_ratio float64 | only_instruct_ratio float64 | resume_from_checkpoint null | init_from_checkpoint string | learning_rate float64 | weight_decay float64 | max_grad_norm float64 | steps int64 | seed int64 | lr_scheduler_type string | warmup_type string | warmup_ratio float64 | warmup_steps int64 | batch_tokens int64 | gradient_accumulation_steps int64 | num_workers int64 | mixed_precision string | allow_tf32 bool | use_deepspeed bool | deepspeed_config null | attn_implementation string | max_sample_tokens int64 | min_sample_tokens int64 | max_batch_size int64 | logging_steps int64 | eval_steps int64 | save_steps int64 | keep_last_n_checkpoints int64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
exp/omnivoice_diac | config/data_config_diac.json | Qwen/Qwen3-0.6B | 1,025 | 1,024 | 8 | [
8,
8,
6,
6,
4,
4,
2,
2
] | 0.1 | [
0,
0.3
] | [
0,
1
] | 1 | 0 | 0 | 0 | null | exp/omnivoice_egy/checkpoint-20000 | 0.00003 | 0.01 | 1 | 20,000 | 42 | cosine | ratio | 0.01 | 0 | 16,384 | 1 | 4 | bf16 | true | false | null | flex_attention | 2,000 | 50 | 64 | 50 | 500 | 1,000 | 5 |
No dataset card yet
- Downloads last month
- 294