Rocketknight1 HF staff commited on
Commit
9a1d1e6
1 Parent(s): c21248e

Upload FalconForCausalLM

Browse files
Files changed (2) hide show
  1. config.json +6 -6
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "./falcon-40b",
3
  "alibi": false,
4
  "apply_residual_connection_post_layernorm": false,
5
  "architectures": [
@@ -10,17 +10,17 @@
10
  "bos_token_id": 11,
11
  "eos_token_id": 11,
12
  "hidden_dropout": 0.0,
13
- "hidden_size": 128,
14
  "initializer_range": 0.02,
15
  "layer_norm_epsilon": 1e-05,
16
  "model_type": "falcon",
17
  "multi_query": true,
18
- "num_attention_heads": 16,
19
- "num_kv_heads": 8,
20
- "num_hidden_layers": 2,
21
  "new_decoder_architecture": true,
 
 
 
22
  "parallel_attn": true,
23
- "torch_dtype": "float32",
24
  "transformers_version": "4.31.0.dev0",
25
  "use_cache": true,
26
  "vocab_size": 65024
 
1
  {
2
+ "_name_or_path": "falcon_convert/repos/falcon-40b",
3
  "alibi": false,
4
  "apply_residual_connection_post_layernorm": false,
5
  "architectures": [
 
10
  "bos_token_id": 11,
11
  "eos_token_id": 11,
12
  "hidden_dropout": 0.0,
13
+ "hidden_size": 1024,
14
  "initializer_range": 0.02,
15
  "layer_norm_epsilon": 1e-05,
16
  "model_type": "falcon",
17
  "multi_query": true,
 
 
 
18
  "new_decoder_architecture": true,
19
+ "num_attention_heads": 128,
20
+ "num_hidden_layers": 2,
21
+ "num_kv_heads": 8,
22
  "parallel_attn": true,
23
+ "torch_dtype": "bfloat16",
24
  "transformers_version": "4.31.0.dev0",
25
  "use_cache": true,
26
  "vocab_size": 65024
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f1ddb10bd43a5bb21cf8c8ce8081f2e6e3645e7d840a103f0cbd425f6560f118
3
- size 34745258
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f04ea3000d7a72b25a7a25d361e5350d1e5a7c99c86e025dc9100bfe95148d9e
3
+ size 175663082