| { |
| "architectures": [ |
| "Qwen2TSForCausalLM" |
| ], |
| "attention_dropout": 0.0, |
| "auto_map": { |
| "AutoConfig": "configuration_qwen2.Qwen2TSConfig", |
| "AutoModel": "modeling_qwen2.Qwen2TSForCausalLM", |
| "AutoModelForCausalLM": "modeling_qwen2.Qwen2TSForCausalLM", |
| "AutoProcessor": "processing_qwen2_ts.Qwen2TSProcessor" |
| }, |
| "bos_token_id": 151643, |
| "eos_token_id": 151645, |
| "hidden_act": "silu", |
| "hidden_size": 5120, |
| "ignore_index": -100, |
| "initializer_range": 0.02, |
| "intermediate_size": 13824, |
| "max_position_embeddings": 32768, |
| "max_window_layers": 70, |
| "model_type": "chatts", |
| "num_attention_heads": 40, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 8, |
| "pad_token_id": 151643, |
| "rms_norm_eps": 1e-06, |
| "rope_theta": 1000000.0, |
| "sliding_window": 131072, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.52.4", |
| "ts": { |
| "embedding_dim": 16, |
| "hidden_size": 5120, |
| "max_length": 32768, |
| "max_sequence_length": 32768, |
| "num_features": 2, |
| "num_layers": 5, |
| "patch_size": 8, |
| "use_position_embedding": true, |
| "use_position_idx": false |
| }, |
| "ts_token_end_index": 151666, |
| "ts_token_start_index": 151665, |
| "use_cache": false, |
| "use_sliding_window": false, |
| "vocab_size": 152064 |
| } |
|
|