| { |
| "architectures": [ |
| "Mamba2ForCausalLM" |
| ], |
| "bos_token_id": 6, |
| "chunk_size": 256, |
| "conv_kernel": 4, |
| "dtype": "float32", |
| "eos_token_id": 6, |
| "expand": 1.5, |
| "head_dim": 64, |
| "hidden_act": "silu", |
| "hidden_size": 512, |
| "initializer_range": 0.1, |
| "layer_norm_epsilon": 1e-05, |
| "model_type": "mamba2", |
| "n_groups": 2, |
| "num_heads": 12, |
| "num_hidden_layers": 9, |
| "pad_token_id": 7, |
| "rescale_prenorm_residual": false, |
| "residual_in_fp32": true, |
| "rms_norm": true, |
| "state_size": 64, |
| "tie_word_embeddings": false, |
| "time_step_floor": 0.0001, |
| "time_step_limit": [ |
| 0.0, |
| { |
| "__float__": "Infinity" |
| } |
| ], |
| "time_step_max": 0.1, |
| "time_step_min": 0.001, |
| "time_step_rank": 32, |
| "transformers_version": "5.3.0", |
| "use_bias": false, |
| "use_cache": false, |
| "use_conv_bias": true, |
| "vocab_size": 8192 |
| } |
|
|