Hugging Face's logo

nm-testing
/
tinyllama-oneshot-w8a8-static-v2

Text Generation
Transformers
Safetensors
llama
conversational
text-generation-inference
Inference Endpoints
8-bit precision
compressed-tensors
Model card Files Community
sadkins65 commited on
Commit
95a9d73
•
1 Parent(s): e47251f

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. config.json +37 -39
  2. model.safetensors +2 -2
config.json CHANGED
@@ -7,48 +7,46 @@
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 1,
9
  "compression_config": {
10
- "quantization_config": {
11
- "config_groups": {
12
- "group_0": {
13
- "input_activations": {
14
- "block_structure": null,
15
- "dynamic": false,
16
- "group_size": null,
17
- "num_bits": 8,
18
- "observer": "minmax",
19
- "observer_kwargs": {},
20
- "strategy": "tensor",
21
- "symmetric": true,
22
- "type": "int"
23
- },
24
- "output_activations": null,
25
- "targets": [
26
- "Linear"
27
- ],
28
- "weights": {
29
- "block_structure": null,
30
- "dynamic": false,
31
- "group_size": null,
32
- "num_bits": 8,
33
- "observer": "minmax",
34
- "observer_kwargs": {},
35
- "strategy": "tensor",
36
- "symmetric": true,
37
- "type": "int"
38
- }
39
  }
40
- },
41
- "format": "int-quantized",
42
- "global_compression_ratio": 1.4583332397611055,
43
- "ignore": [
44
- "lm_head"
45
- ],
46
- "quant_method": "sparseml",
47
- "quantization_status": "frozen"
48
  },
 
 
 
 
 
 
 
49
  "sparsity_config": {
50
  "format": "dense",
51
- "global_sparsity": 7.819636413374341,
52
  "registry_requires_subclass": false,
53
  "sparsity_structure": "unstructured"
54
  }
@@ -68,7 +66,7 @@
68
  "rope_scaling": null,
69
  "rope_theta": 10000.0,
70
  "tie_word_embeddings": false,
71
- "torch_dtype": "bfloat16",
72
  "transformers_version": "4.40.0",
73
  "use_cache": true,
74
  "vocab_size": 32000
 
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 1,
9
  "compression_config": {
10
+ "config_groups": {
11
+ "group_0": {
12
+ "input_activations": {
13
+ "block_structure": null,
14
+ "dynamic": false,
15
+ "group_size": null,
16
+ "num_bits": 8,
17
+ "observer": "minmax",
18
+ "observer_kwargs": {},
19
+ "strategy": "tensor",
20
+ "symmetric": true,
21
+ "type": "int"
22
+ },
23
+ "output_activations": null,
24
+ "targets": [
25
+ "Linear"
26
+ ],
27
+ "weights": {
28
+ "block_structure": null,
29
+ "dynamic": false,
30
+ "group_size": null,
31
+ "num_bits": 8,
32
+ "observer": "minmax",
33
+ "observer_kwargs": {},
34
+ "strategy": "tensor",
35
+ "symmetric": true,
36
+ "type": "int"
 
 
37
  }
38
+ }
 
 
 
 
 
 
 
39
  },
40
+ "format": "int-quantized",
41
+ "global_compression_ratio": 1.4583332397611055,
42
+ "ignore": [
43
+ "lm_head"
44
+ ],
45
+ "quant_method": "compressed-tensors",
46
+ "quantization_status": "frozen",
47
  "sparsity_config": {
48
  "format": "dense",
49
+ "global_sparsity": 7.8259900429979625,
50
  "registry_requires_subclass": false,
51
  "sparsity_structure": "unstructured"
52
  }
 
66
  "rope_scaling": null,
67
  "rope_theta": 10000.0,
68
  "tie_word_embeddings": false,
69
+ "torch_dtype": "float16",
70
  "transformers_version": "4.40.0",
71
  "use_cache": true,
72
  "vocab_size": 32000
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e2fe1d8ccc00820db4f20dd2ff4deefe82f1e01794e46fc1c44e1579a417172a
3
- size 1231269800
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85e1306ddfc0991bfe20892458b4caecd4605f2f02faabee21a71e211c177a28
3
+ size 1231269448