tf3-50m-base-mamba / config.json
andreiPiscoran's picture
Upload 6 files
dcde50e verified
raw
history blame contribute delete
759 Bytes
{
"architectures": [
"MambaForCausalLM"
],
"bos_token_id": 2,
"conv_kernel": 4,
"dtype": "float32",
"eos_token_id": 3,
"expand": 2,
"hidden_act": "silu",
"hidden_size": 512,
"initializer_range": 0.1,
"intermediate_size": 1024,
"layer_norm_epsilon": 1e-05,
"model_type": "mamba",
"num_hidden_layers": 18,
"pad_token_id": 0,
"rescale_prenorm_residual": false,
"residual_in_fp32": true,
"state_size": 64,
"time_step_floor": 0.0001,
"time_step_init_scheme": "random",
"time_step_max": 0.1,
"time_step_min": 0.001,
"time_step_rank": 32,
"time_step_scale": 1.0,
"transformers_version": "4.56.1",
"use_bias": false,
"use_cache": true,
"use_conv_bias": true,
"use_mambapy": false,
"vocab_size": 32000
}