初始化项目,由ModelHub XC社区提供模型
Model: Vcecca/mamba-50m Source: Original Platform
This commit is contained in:
34
config.json
Normal file
34
config.json
Normal file
@@ -0,0 +1,34 @@
|
||||
{
|
||||
"_name_or_path": "./checkpoints/snapshot-articles-3180018",
|
||||
"architectures": [
|
||||
"MambaForCausalLM"
|
||||
],
|
||||
"bos_token_id": 0,
|
||||
"conv_kernel": 4,
|
||||
"eos_token_id": 0,
|
||||
"expand": 2,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 384,
|
||||
"initializer_range": 0.1,
|
||||
"intermediate_size": 768,
|
||||
"layer_norm_epsilon": 1e-05,
|
||||
"model_type": "mamba",
|
||||
"num_hidden_layers": 32,
|
||||
"pad_token_id": 0,
|
||||
"rescale_prenorm_residual": false,
|
||||
"residual_in_fp32": true,
|
||||
"state_size": 16,
|
||||
"time_step_floor": 0.0001,
|
||||
"time_step_init_scheme": "random",
|
||||
"time_step_max": 0.1,
|
||||
"time_step_min": 0.001,
|
||||
"time_step_rank": 24,
|
||||
"time_step_scale": 1.0,
|
||||
"torch_dtype": "float32",
|
||||
"transformers_version": "4.48.0",
|
||||
"use_bias": false,
|
||||
"use_cache": false,
|
||||
"use_conv_bias": true,
|
||||
"use_mambapy": false,
|
||||
"vocab_size": 50280
|
||||
}
|
||||
Reference in New Issue
Block a user