qgallouedec/tiny-FalconMambaForCausalLM
0127
1{2 "architectures": [3 "FalconMambaForCausalLM"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": 8,11 "initializer_range": 0.1,12 "intermediate_size": 32,13 "layer_norm_epsilon": 1e-05,14 "mixer_rms_eps": 1e-06,15 "model_type": "falcon_mamba",16 "num_attention_heads": 4,17 "num_hidden_layers": 2,18 "num_key_value_heads": 2,19 "pad_token_id": 0,20 "rescale_prenorm_residual": false,21 "residual_in_fp32": true,22 "state_size": 16,23 "time_step_floor": 0.0001,24 "time_step_init_scheme": "random",25 "time_step_max": 0.1,26 "time_step_min": 0.001,27 "time_step_rank": 1,28 "time_step_scale": 1.0,29 "torch_dtype": "float32",30 "transformers_version": "4.47.0.dev0",31 "use_bias": false,32 "use_cache": true,33 "use_conv_bias": true,34 "use_mambapy": false,35 "vocab_size": 6503636}37 