Hamilton / configs /mamba_1B.json
ZhenbinWang's picture
Upload 25 files
9fb98d3 verified
raw
history blame contribute delete
711 Bytes
{
"bos_token_id": 1,
"conv_kernel": 4,
"eos_token_id": 2,
"expand": 2,
"fuse_cross_entropy": true,
"fuse_norm": true,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"model_type": "mamba",
"norm_eps": 1e-05,
"num_hidden_layers": 48,
"pad_token_id": 0,
"rescale_prenorm_residual": false,
"residual_in_fp32": false,
"state_size": 16,
"tie_word_embeddings": false,
"time_step_floor": 0.0001,
"time_step_init_scheme": "random",
"time_step_max": 0.1,
"time_step_min": 0.001,
"time_step_rank": 128,
"time_step_scale": 1.0,
"transformers_version": "4.50.1",
"use_bias": false,
"use_cache": true,
"use_conv_bias": true,
"vocab_size": 32000
}