w601sxs commited on
Commit
336f308
·
1 Parent(s): bee07a2

Upload GPTNeoXForCausalLM

Browse files
Files changed (3) hide show
  1. config.json +4 -1
  2. generation_config.json +1 -1
  3. pytorch_model.bin +2 -2
config.json CHANGED
@@ -3,10 +3,12 @@
3
  "architectures": [
4
  "GPTNeoXForCausalLM"
5
  ],
 
6
  "bos_token_id": 0,
7
  "classifier_dropout": 0.1,
8
  "eos_token_id": 0,
9
  "hidden_act": "gelu",
 
10
  "hidden_size": 2048,
11
  "initializer_range": 0.02,
12
  "intermediate_size": 8192,
@@ -15,11 +17,12 @@
15
  "model_type": "gpt_neox",
16
  "num_attention_heads": 8,
17
  "num_hidden_layers": 16,
 
18
  "rotary_emb_base": 10000,
19
  "rotary_pct": 0.25,
20
  "tie_word_embeddings": false,
21
  "torch_dtype": "bfloat16",
22
- "transformers_version": "4.30.2",
23
  "use_cache": true,
24
  "use_parallel_residual": true,
25
  "vocab_size": 50304
 
3
  "architectures": [
4
  "GPTNeoXForCausalLM"
5
  ],
6
+ "attention_dropout": 0.0,
7
  "bos_token_id": 0,
8
  "classifier_dropout": 0.1,
9
  "eos_token_id": 0,
10
  "hidden_act": "gelu",
11
+ "hidden_dropout": 0.0,
12
  "hidden_size": 2048,
13
  "initializer_range": 0.02,
14
  "intermediate_size": 8192,
 
17
  "model_type": "gpt_neox",
18
  "num_attention_heads": 8,
19
  "num_hidden_layers": 16,
20
+ "rope_scaling": null,
21
  "rotary_emb_base": 10000,
22
  "rotary_pct": 0.25,
23
  "tie_word_embeddings": false,
24
  "torch_dtype": "bfloat16",
25
+ "transformers_version": "4.34.0",
26
  "use_cache": true,
27
  "use_parallel_residual": true,
28
  "vocab_size": 50304
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
- "transformers_version": "4.30.2"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 0,
4
  "eos_token_id": 0,
5
+ "transformers_version": "4.34.0"
6
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed27803294202b8677f80ea816ad994fa9b95b6af411e7370931d192cb24baf4
3
- size 2090756445
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a14426fd746b086ef58d7bce4ac74c93e391b86894c220cd8f333474dd76418b
3
+ size 2023636829