osbm commited on
Commit
49ea673
1 Parent(s): 3582e51

Upload 3 files

Browse files
Files changed (3) hide show
  1. config.json +31 -0
  2. quant_config.json +11 -0
  3. quantized_weights.pt +3 -0
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/opt-6.7b",
3
+ "_remove_final_layer_norm": false,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
6
+ "architectures": [
7
+ "OPTForCausalLM"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "bos_token_id": 2,
11
+ "do_layer_norm_before": true,
12
+ "dropout": 0.1,
13
+ "enable_bias": true,
14
+ "eos_token_id": 2,
15
+ "ffn_dim": 16384,
16
+ "hidden_size": 4096,
17
+ "init_std": 0.02,
18
+ "layer_norm_elementwise_affine": true,
19
+ "layerdrop": 0.0,
20
+ "max_position_embeddings": 2048,
21
+ "model_type": "opt",
22
+ "num_attention_heads": 32,
23
+ "num_hidden_layers": 32,
24
+ "pad_token_id": 1,
25
+ "prefix": "</s>",
26
+ "torch_dtype": "float16",
27
+ "transformers_version": "4.34.0",
28
+ "use_cache": true,
29
+ "vocab_size": 50272,
30
+ "word_embed_proj_dim": 4096
31
+ }
quant_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": "c4",
3
+ "bits": 4,
4
+ "nsamples": 128,
5
+ "groupsize": 64,
6
+ "act_order": true,
7
+ "percdamp": 0.01,
8
+ "seed": 0,
9
+ "nearest": false,
10
+ "save": "./opt-6.7b-4bit-quantized"
11
+ }
quantized_weights.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3d62fc95c7ed5b86f1ca3652aa1c504bc762da3660f9e2373489fee1d76a4ca
3
+ size 3910030998