breadlicker45 commited on
Commit
78619df
1 Parent(s): 4b62aef

Upload 3 files

Browse files
Files changed (3) hide show
  1. config.json +6 -6
  2. generation_config.json +6 -0
  3. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,25 +1,25 @@
1
  {
2
- "_name_or_path": "EleutherAI/pythia-19m",
3
  "architectures": [
4
  "GPTNeoXForCausalLM"
5
  ],
6
  "bos_token_id": 0,
7
  "eos_token_id": 0,
8
  "hidden_act": "gelu",
9
- "hidden_size": 512,
10
  "initializer_range": 0.02,
11
- "intermediate_size": 2048,
12
  "layer_norm_eps": 1e-05,
13
  "line_by_line": true,
14
  "max_position_embeddings": 2048,
15
  "model_type": "gpt_neox",
16
- "num_attention_heads": 8,
17
- "num_hidden_layers": 6,
18
  "rotary_emb_base": 10000,
19
  "rotary_pct": 0.25,
20
  "tie_word_embeddings": false,
21
  "torch_dtype": "float32",
22
- "transformers_version": "4.25.1",
23
  "use_cache": true,
24
  "use_parallel_residual": true,
25
  "vocab_size": 50304
 
1
  {
2
+ "_name_or_path": "EleutherAI/pythia-160m",
3
  "architectures": [
4
  "GPTNeoXForCausalLM"
5
  ],
6
  "bos_token_id": 0,
7
  "eos_token_id": 0,
8
  "hidden_act": "gelu",
9
+ "hidden_size": 768,
10
  "initializer_range": 0.02,
11
+ "intermediate_size": 3072,
12
  "layer_norm_eps": 1e-05,
13
  "line_by_line": true,
14
  "max_position_embeddings": 2048,
15
  "model_type": "gpt_neox",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 12,
18
  "rotary_emb_base": 10000,
19
  "rotary_pct": 0.25,
20
  "tie_word_embeddings": false,
21
  "torch_dtype": "float32",
22
+ "transformers_version": "4.26.1",
23
  "use_cache": true,
24
  "use_parallel_residual": true,
25
  "vocab_size": 50304
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "transformers_version": "4.26.1"
6
+ }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d946f5fcb5a37f0d3623d61061b22e26b2f92fdc7e2d430434be567ce32a1ce6
3
- size 25915
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2f9b98697e10ed5caa675082caf7983c4f756f25ceec65bda7f0ea4e6e76128
3
+ size 699682733