Hugging Face's logo

steerapi
/
Llama-2-7b-chat-hf-onnx-awq

Text Generation
Transformers
ONNX
llama
Model card Files Community
1
steerapi commited on
Commit
f236c3f
·
1 Parent(s): 39226cd

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. decoder_model.onnx +0 -3
  2. decoder_model.onnx_data +0 -3
  3. decoder_model_merged.onnx +0 -3
  4. decoder_model_merged.onnx_data +0 -3
  5. decoder_model_merged_quantized.onnx +0 -3
  6. decoder_model_merged_quantized.onnx_data +0 -3
  7. decoder_with_past_model.onnx +0 -3
  8. decoder_with_past_model.onnx_data +0 -3
  9. onnx/decoder_model_merged_quantized.onnx +2 -2
  10. onnx/decoder_model_merged_quantized.onnx_data +1 -1
  11. {q1 → onnx/q1}/decoder_model_merged_quantized.onnx +0 -0
  12. {q1 → onnx/q1}/decoder_model_merged_quantized.onnx_data +0 -0
  13. onnx/quantize_config.json +41 -0
  14. quantize_config.json +21 -21
decoder_model.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:d24bcccc1112825def21cf4241933aae2c89187b3663928e9a083faeaed92249
3
- size 5437678
 
 
 
 
decoder_model.onnx_data DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:58e859c15c13e65b9a5478e3a4e3dc81f5b0d5500d5d18c942064a447f6af150
3
- size 26953662464
 
 
 
 
decoder_model_merged.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:48a4703bff0b9eff305ece6e15f20ab52d52d48eccaa28bbb2159f38cebeacc4
3
- size 10888493
 
 
 
 
decoder_model_merged.onnx_data DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:58e859c15c13e65b9a5478e3a4e3dc81f5b0d5500d5d18c942064a447f6af150
3
- size 26953662464
 
 
 
 
decoder_model_merged_quantized.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f1266c653f7026aa4126cac8eadae4eb82e617e37ed65cd60bfda5a490dcd305
3
- size 19045741
 
 
 
 
decoder_model_merged_quantized.onnx_data DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:0f57f92dce820666b27faace4a325a7d8d41691b4f9d3e3540ba70e696b42a06
3
- size 6739214336
 
 
 
 
decoder_with_past_model.onnx DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4425410e11031c394eb76701bc8c82c0fdf74273934e62d6028d9d3242a67ce7
3
- size 5471136
 
 
 
 
decoder_with_past_model.onnx_data DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:58e859c15c13e65b9a5478e3a4e3dc81f5b0d5500d5d18c942064a447f6af150
3
- size 26953662464
 
 
 
 
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:97ef03e1d46da5972d3324761e9b16a43112bb48a7eaa62658d2a4e2141b370f
3
- size 19048364
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1266c653f7026aa4126cac8eadae4eb82e617e37ed65cd60bfda5a490dcd305
3
+ size 19045741
onnx/decoder_model_merged_quantized.onnx_data CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dab778698ac3bc754db6291ecc0c43d92278f6a69a7d5d006e39a4e98349c2b8
3
  size 6739214336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f57f92dce820666b27faace4a325a7d8d41691b4f9d3e3540ba70e696b42a06
3
  size 6739214336
{q1 → onnx/q1}/decoder_model_merged_quantized.onnx RENAMED
File without changes
{q1 → onnx/q1}/decoder_model_merged_quantized.onnx_data RENAMED
File without changes
onnx/quantize_config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "use_external_data_format": true,
5
+ "per_model_config": {
6
+ "decoder_model_merged": {
7
+ "op_types": [
8
+ "Concat",
9
+ "Transpose",
10
+ "Neg",
11
+ "Sigmoid",
12
+ "ReduceMean",
13
+ "Identity",
14
+ "Expand",
15
+ "Shape",
16
+ "Equal",
17
+ "Sqrt",
18
+ "Range",
19
+ "Where",
20
+ "Gather",
21
+ "Pow",
22
+ "Constant",
23
+ "Sub",
24
+ "Less",
25
+ "Div",
26
+ "MatMul",
27
+ "Cast",
28
+ "Unsqueeze",
29
+ "If",
30
+ "Softmax",
31
+ "Add",
32
+ "ConstantOfShape",
33
+ "Squeeze",
34
+ "Mul",
35
+ "Reshape",
36
+ "Slice"
37
+ ],
38
+ "weight_type": "QInt8"
39
+ }
40
+ }
41
+ }
quantize_config.json CHANGED
@@ -5,34 +5,34 @@
5
  "per_model_config": {
6
  "decoder_model_merged": {
7
  "op_types": [
8
- "Concat",
9
- "Transpose",
10
- "Neg",
11
  "Sigmoid",
12
- "ReduceMean",
13
- "Identity",
14
- "Expand",
15
- "Shape",
16
- "Equal",
17
- "Sqrt",
18
- "Range",
19
- "Where",
20
- "Gather",
21
- "Pow",
22
- "Constant",
23
- "Sub",
24
- "Less",
25
  "Div",
26
  "MatMul",
27
  "Cast",
 
 
 
 
 
 
 
 
 
 
 
 
28
  "Unsqueeze",
29
- "If",
 
 
30
  "Softmax",
31
- "Add",
32
- "ConstantOfShape",
 
33
  "Squeeze",
34
- "Mul",
35
- "Reshape",
36
  "Slice"
37
  ],
38
  "weight_type": "QInt8"
 
5
  "per_model_config": {
6
  "decoder_model_merged": {
7
  "op_types": [
 
 
 
8
  "Sigmoid",
9
+ "If",
10
+ "Reshape",
 
 
 
 
 
 
 
 
 
 
 
11
  "Div",
12
  "MatMul",
13
  "Cast",
14
+ "ConstantOfShape",
15
+ "Constant",
16
+ "Neg",
17
+ "Gather",
18
+ "Mul",
19
+ "Add",
20
+ "Where",
21
+ "Less",
22
+ "Expand",
23
+ "Concat",
24
+ "Sub",
25
+ "Equal",
26
  "Unsqueeze",
27
+ "Pow",
28
+ "Sqrt",
29
+ "Transpose",
30
  "Softmax",
31
+ "Identity",
32
+ "Range",
33
+ "Shape",
34
  "Squeeze",
35
+ "ReduceMean",
 
36
  "Slice"
37
  ],
38
  "weight_type": "QInt8"