Upload MambaForCausalLM
Browse files- config.json +7 -2
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -5,21 +5,26 @@
|
|
| 5 |
],
|
| 6 |
"bos_token_id": 1,
|
| 7 |
"conv_kernel": 4,
|
|
|
|
| 8 |
"eos_token_id": 2,
|
| 9 |
"expand": 2,
|
| 10 |
"fused_add_norm": true,
|
| 11 |
"hidden_act": "silu",
|
| 12 |
-
"num_hidden_layers": 48,
|
| 13 |
"hidden_size": 2048,
|
| 14 |
"initializer_range": 0.1,
|
|
|
|
| 15 |
"layer_norm_epsilon": 1e-05,
|
| 16 |
"model_type": "mamba",
|
| 17 |
-
"
|
|
|
|
|
|
|
| 18 |
"residual_in_fp32": true,
|
| 19 |
"rms_norm": true,
|
|
|
|
| 20 |
"state_size": 16,
|
| 21 |
"time_step_rank": 128,
|
| 22 |
"torch_dtype": "float16",
|
|
|
|
| 23 |
"use_bias": false,
|
| 24 |
"use_cache": true,
|
| 25 |
"use_conv_bias": true,
|
|
|
|
| 5 |
],
|
| 6 |
"bos_token_id": 1,
|
| 7 |
"conv_kernel": 4,
|
| 8 |
+
"d_model": 2048,
|
| 9 |
"eos_token_id": 2,
|
| 10 |
"expand": 2,
|
| 11 |
"fused_add_norm": true,
|
| 12 |
"hidden_act": "silu",
|
|
|
|
| 13 |
"hidden_size": 2048,
|
| 14 |
"initializer_range": 0.1,
|
| 15 |
+
"intermediate_size": 4096,
|
| 16 |
"layer_norm_epsilon": 1e-05,
|
| 17 |
"model_type": "mamba",
|
| 18 |
+
"n_layer": 48,
|
| 19 |
+
"num_hidden_layers": 48,
|
| 20 |
+
"pad_vocab_size_multiple": 8,
|
| 21 |
"residual_in_fp32": true,
|
| 22 |
"rms_norm": true,
|
| 23 |
+
"ssm_cfg": {},
|
| 24 |
"state_size": 16,
|
| 25 |
"time_step_rank": 128,
|
| 26 |
"torch_dtype": "float16",
|
| 27 |
+
"transformers_version": "4.39.0.dev0",
|
| 28 |
"use_bias": false,
|
| 29 |
"use_cache": true,
|
| 30 |
"use_conv_bias": true,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed63a38777f21d233810c6cb62dd57f145c8ee7260bcb919329cef545d41adb1
|
| 3 |
+
size 2750701096
|