Mamba-v0
Browse files
    	
        chess-mamba-v0/config.json
    ADDED
    
    | 
         @@ -0,0 +1,32 @@ 
     | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
| 
         | 
|
| 1 | 
         
            +
            {
         
     | 
| 2 | 
         
            +
              "architectures": [
         
     | 
| 3 | 
         
            +
                "ChessMambaModel"
         
     | 
| 4 | 
         
            +
              ],
         
     | 
| 5 | 
         
            +
              "bos_token_id": 0,
         
     | 
| 6 | 
         
            +
              "conv_kernel": 4,
         
     | 
| 7 | 
         
            +
              "eos_token_id": 0,
         
     | 
| 8 | 
         
            +
              "expand": 2,
         
     | 
| 9 | 
         
            +
              "hidden_act": "silu",
         
     | 
| 10 | 
         
            +
              "hidden_size": 512,
         
     | 
| 11 | 
         
            +
              "initializer_range": 0.1,
         
     | 
| 12 | 
         
            +
              "intermediate_size": 1024,
         
     | 
| 13 | 
         
            +
              "layer_norm_epsilon": 1e-05,
         
     | 
| 14 | 
         
            +
              "model_type": "mamba",
         
     | 
| 15 | 
         
            +
              "num_hidden_layers": 16,
         
     | 
| 16 | 
         
            +
              "pad_token_id": 0,
         
     | 
| 17 | 
         
            +
              "rescale_prenorm_residual": false,
         
     | 
| 18 | 
         
            +
              "residual_in_fp32": true,
         
     | 
| 19 | 
         
            +
              "state_size": 32,
         
     | 
| 20 | 
         
            +
              "time_step_floor": 0.0001,
         
     | 
| 21 | 
         
            +
              "time_step_init_scheme": "random",
         
     | 
| 22 | 
         
            +
              "time_step_max": 0.1,
         
     | 
| 23 | 
         
            +
              "time_step_min": 0.001,
         
     | 
| 24 | 
         
            +
              "time_step_rank": 48,
         
     | 
| 25 | 
         
            +
              "time_step_scale": 1.0,
         
     | 
| 26 | 
         
            +
              "torch_dtype": "float32",
         
     | 
| 27 | 
         
            +
              "transformers_version": "4.39.0.dev0",
         
     | 
| 28 | 
         
            +
              "use_bias": false,
         
     | 
| 29 | 
         
            +
              "use_cache": true,
         
     | 
| 30 | 
         
            +
              "use_conv_bias": true,
         
     | 
| 31 | 
         
            +
              "vocab_size": 104
         
     | 
| 32 | 
         
            +
            }
         
     | 
    	
        chess-mamba-v0/generation_config.json
    ADDED
    
    | 
         @@ -0,0 +1,7 @@ 
     | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
| 
         | 
|
| 1 | 
         
            +
            {
         
     | 
| 2 | 
         
            +
              "_from_model_config": true,
         
     | 
| 3 | 
         
            +
              "bos_token_id": 0,
         
     | 
| 4 | 
         
            +
              "eos_token_id": 0,
         
     | 
| 5 | 
         
            +
              "pad_token_id": 0,
         
     | 
| 6 | 
         
            +
              "transformers_version": "4.39.0.dev0"
         
     | 
| 7 | 
         
            +
            }
         
     | 
    	
        chess-mamba-v0/model.safetensors
    ADDED
    
    | 
         @@ -0,0 +1,3 @@ 
     | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
| 
         | 
|
| 1 | 
         
            +
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:d293c5adbd320f434128c9e4e4027dfebf4a97c16789e7a70a8e6385c53b2455
         
     | 
| 3 | 
         
            +
            size 113969896
         
     |