qgallouedec HF Staff commited on
Commit
148d3c8
·
verified ·
1 Parent(s): 88e0b1b

Upload FalconMambaForCausalLM

Browse files
Files changed (3) hide show
  1. config.json +4 -4
  2. generation_config.json +1 -1
  3. model.safetensors +2 -2
config.json CHANGED
@@ -4,12 +4,13 @@
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": 16,
13
  "layer_norm_epsilon": 1e-05,
14
  "mixer_rms_eps": 1e-06,
15
  "model_type": "falcon_mamba",
@@ -26,11 +27,10 @@
26
  "time_step_min": 0.001,
27
  "time_step_rank": 1,
28
  "time_step_scale": 1.0,
29
- "torch_dtype": "float32",
30
- "transformers_version": "4.55.0.dev0",
31
  "use_bias": false,
32
  "use_cache": true,
33
  "use_conv_bias": true,
34
  "use_falcon_mambapy": false,
35
- "vocab_size": 65036
36
  }
 
4
  ],
5
  "bos_token_id": 0,
6
  "conv_kernel": 4,
7
+ "dtype": "float32",
8
  "eos_token_id": 0,
9
  "expand": 2,
10
  "hidden_act": "silu",
11
  "hidden_size": 8,
12
  "initializer_range": 0.1,
13
+ "intermediate_size": 32,
14
  "layer_norm_epsilon": 1e-05,
15
  "mixer_rms_eps": 1e-06,
16
  "model_type": "falcon_mamba",
 
27
  "time_step_min": 0.001,
28
  "time_step_rank": 1,
29
  "time_step_scale": 1.0,
30
+ "transformers_version": "4.57.0.dev0",
 
31
  "use_bias": false,
32
  "use_cache": true,
33
  "use_conv_bias": true,
34
  "use_falcon_mambapy": false,
35
+ "vocab_size": 65024
36
  }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
  "pad_token_id": 0,
6
- "transformers_version": "4.55.0.dev0"
7
  }
 
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
  "pad_token_id": 0,
6
+ "transformers_version": "4.57.0.dev0"
7
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8815c8a41013afd78473fa748fd03a91bad7206057bb412300ec7782563317fa
3
- size 2093848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f424c99dc84cd2e8d1e562dfed99bb6744c1b7d05d1d210204073513bdc74cff
3
+ size 2103832