sharpenb commited on
Commit
3bdb7d6
1 Parent(s): 4fd8b53

Upload folder using huggingface_hub (#3)

Browse files

- 5a9c1b2a9a43e2cf12fc0c12abbfac6695bfda7b5183d5abc61390d80036790e (88d91b0d830871931d727068a35ad3b3be144c3b)
- 1a3cf829432d35b5ab5ff7e326a8db49db325db7a6451985f8806332e6a37529 (fa4ab1a44fc2fcdc84025de924b3942e1134d967)
- 98ba3df194d2ac1bb413b28d0618a6fddc17cc2377d1793229b8a4dc005864e4 (dbf24e7e3458854369ff9e0986339fe26a804ca8)

Files changed (4) hide show
  1. README.md +2 -2
  2. config.json +2 -2
  3. plots.png +0 -0
  4. smash_config.json +1 -1
README.md CHANGED
@@ -34,7 +34,7 @@ tags:
34
 
35
  ## Results
36
 
37
- Detailed efficiency metrics coming soon!
38
 
39
  **Frequently Asked Questions**
40
  - ***How does the compression work?*** The model is compressed with llm-int8.
@@ -61,7 +61,7 @@ You can run the smashed model with these steps:
61
  from transformers import AutoModelForCausalLM, AutoTokenizer
62
 
63
  model = AutoModelForCausalLM.from_pretrained("PrunaAI/GritLM-GritLM-7B-bnb-8bit-smashed",
64
- trust_remote_code=True)
65
  tokenizer = AutoTokenizer.from_pretrained("GritLM/GritLM-7B")
66
 
67
  input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
 
34
 
35
  ## Results
36
 
37
+ ![image info](./plots.png)
38
 
39
  **Frequently Asked Questions**
40
  - ***How does the compression work?*** The model is compressed with llm-int8.
 
61
  from transformers import AutoModelForCausalLM, AutoTokenizer
62
 
63
  model = AutoModelForCausalLM.from_pretrained("PrunaAI/GritLM-GritLM-7B-bnb-8bit-smashed",
64
+ trust_remote_code=True, device_map='auto')
65
  tokenizer = AutoTokenizer.from_pretrained("GritLM/GritLM-7B")
66
 
67
  input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/tmp/tmpkyzzzm49",
3
  "architectures": [
4
  "MistralForCausalLM"
5
  ],
@@ -29,7 +29,7 @@
29
  "quantization_config": {
30
  "bnb_4bit_compute_dtype": "bfloat16",
31
  "bnb_4bit_quant_type": "fp4",
32
- "bnb_4bit_use_double_quant": true,
33
  "llm_int8_enable_fp32_cpu_offload": false,
34
  "llm_int8_has_fp16_weight": false,
35
  "llm_int8_skip_modules": [
 
1
  {
2
+ "_name_or_path": "/tmp/tmp7ro6rtjc",
3
  "architectures": [
4
  "MistralForCausalLM"
5
  ],
 
29
  "quantization_config": {
30
  "bnb_4bit_compute_dtype": "bfloat16",
31
  "bnb_4bit_quant_type": "fp4",
32
+ "bnb_4bit_use_double_quant": false,
33
  "llm_int8_enable_fp32_cpu_offload": false,
34
  "llm_int8_has_fp16_weight": false,
35
  "llm_int8_skip_modules": [
plots.png ADDED
smash_config.json CHANGED
@@ -8,7 +8,7 @@
8
  "compilers": "None",
9
  "task": "text_text_generation",
10
  "device": "cuda",
11
- "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelskz0k30m8",
12
  "batch_size": 1,
13
  "model_name": "GritLM/GritLM-7B",
14
  "pruning_ratio": 0.0,
 
8
  "compilers": "None",
9
  "task": "text_text_generation",
10
  "device": "cuda",
11
+ "cache_dir": "/ceph/hdd/staff/charpent/.cache/models3rsn8hei",
12
  "batch_size": 1,
13
  "model_name": "GritLM/GritLM-7B",
14
  "pruning_ratio": 0.0,