Qubitium commited on
Commit
1810a8c
1 Parent(s): 95d4e99

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +94 -1
README.md CHANGED
@@ -27,7 +27,7 @@ This model has been quantized using [GPTQModel](https://github.com/ModelCloud/GP
27
  - **meta**:
28
  - **quantizer**: "gptqmodel:0.9.9-dev0"
29
 
30
- **Here is an example:**
31
  ```python
32
  from transformers import AutoTokenizer
33
  from gptqmodel import GPTQModel
@@ -45,4 +45,97 @@ outputs = model.generate(input_ids=input_tensor.to(model.device), max_new_tokens
45
  result = tokenizer.decode(outputs[0][input_tensor.shape[1]:], skip_special_tokens=True)
46
 
47
  print(result)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  ```
 
27
  - **meta**:
28
  - **quantizer**: "gptqmodel:0.9.9-dev0"
29
 
30
+ ## Example:
31
  ```python
32
  from transformers import AutoTokenizer
33
  from gptqmodel import GPTQModel
 
45
  result = tokenizer.decode(outputs[0][input_tensor.shape[1]:], skip_special_tokens=True)
46
 
47
  print(result)
48
+ ```
49
+
50
+ ## lm-eval benchmark
51
+
52
+ ```
53
+ | Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
54
+ |---------------------------------------|------:|------|-----:|----------|---|-----:|---|-----:|
55
+ |arc_challenge | 1|none | 0|acc |↑ |0.4889|± |0.0146|
56
+ | | |none | 0|acc_norm |↑ |0.5265|± |0.0146|
57
+ |arc_easy | 1|none | 0|acc |↑ |0.7908|± |0.0083|
58
+ | | |none | 0|acc_norm |↑ |0.7702|± |0.0086|
59
+ |boolq | 2|none | 0|acc |↑ |0.8404|± |0.0064|
60
+ |hellaswag | 1|none | 0|acc |↑ |0.5748|± |0.0049|
61
+ | | |none | 0|acc_norm |↑ |0.7718|± |0.0042|
62
+ |lambada_openai | 1|none | 0|acc |↑ |0.7110|± |0.0063|
63
+ | | |none | 0|perplexity|↓ |4.0554|± |0.0955|
64
+ |mmlu | 1|none | |acc |↑ |0.6411|± |0.0038|
65
+ | - humanities | 1|none | |acc |↑ |0.5896|± |0.0068|
66
+ | - formal_logic | 0|none | 0|acc |↑ |0.4762|± |0.0447|
67
+ | - high_school_european_history | 0|none | 0|acc |↑ |0.7273|± |0.0348|
68
+ | - high_school_us_history | 0|none | 0|acc |↑ |0.8088|± |0.0276|
69
+ | - high_school_world_history | 0|none | 0|acc |↑ |0.8270|± |0.0246|
70
+ | - international_law | 0|none | 0|acc |↑ |0.7851|± |0.0375|
71
+ | - jurisprudence | 0|none | 0|acc |↑ |0.7593|± |0.0413|
72
+ | - logical_fallacies | 0|none | 0|acc |↑ |0.7607|± |0.0335|
73
+ | - moral_disputes | 0|none | 0|acc |↑ |0.7197|± |0.0242|
74
+ | - moral_scenarios | 0|none | 0|acc |↑ |0.4045|± |0.0164|
75
+ | - philosophy | 0|none | 0|acc |↑ |0.7074|± |0.0258|
76
+ | - prehistory | 0|none | 0|acc |↑ |0.6975|± |0.0256|
77
+ | - professional_law | 0|none | 0|acc |↑ |0.4817|± |0.0128|
78
+ | - world_religions | 0|none | 0|acc |↑ |0.7953|± |0.0309|
79
+ | - other | 1|none | |acc |↑ |0.7100|± |0.0078|
80
+ | - business_ethics | 0|none | 0|acc |↑ |0.6700|± |0.0473|
81
+ | - clinical_knowledge | 0|none | 0|acc |↑ |0.7660|± |0.0261|
82
+ | - college_medicine | 0|none | 0|acc |↑ |0.6590|± |0.0361|
83
+ | - global_facts | 0|none | 0|acc |↑ |0.3600|± |0.0482|
84
+ | - human_aging | 0|none | 0|acc |↑ |0.6547|± |0.0319|
85
+ | - management | 0|none | 0|acc |↑ |0.8447|± |0.0359|
86
+ | - marketing | 0|none | 0|acc |↑ |0.8803|± |0.0213|
87
+ | - medical_genetics | 0|none | 0|acc |↑ |0.7100|± |0.0456|
88
+ | - miscellaneous | 0|none | 0|acc |↑ |0.8161|± |0.0139|
89
+ | - nutrition | 0|none | 0|acc |↑ |0.7124|± |0.0259|
90
+ | - professional_accounting | 0|none | 0|acc |↑ |0.4787|± |0.0298|
91
+ | - professional_medicine | 0|none | 0|acc |↑ |0.7279|± |0.0270|
92
+ | - virology | 0|none | 0|acc |↑ |0.5181|± |0.0389|
93
+ | - social sciences | 1|none | |acc |↑ |0.7312|± |0.0078|
94
+ | - econometrics | 0|none | 0|acc |↑ |0.4035|± |0.0462|
95
+ | - high_school_geography | 0|none | 0|acc |↑ |0.8232|± |0.0272|
96
+ | - high_school_government_and_politics| 0|none | 0|acc |↑ |0.8653|± |0.0246|
97
+ | - high_school_macroeconomics | 0|none | 0|acc |↑ |0.6128|± |0.0247|
98
+ | - high_school_microeconomics | 0|none | 0|acc |↑ |0.7227|± |0.0291|
99
+ | - high_school_psychology | 0|none | 0|acc |↑ |0.8422|± |0.0156|
100
+ | - human_sexuality | 0|none | 0|acc |↑ |0.7634|± |0.0373|
101
+ | - professional_psychology | 0|none | 0|acc |↑ |0.6585|± |0.0192|
102
+ | - public_relations | 0|none | 0|acc |↑ |0.6182|± |0.0465|
103
+ | - security_studies | 0|none | 0|acc |↑ |0.7306|± |0.0284|
104
+ | - sociology | 0|none | 0|acc |↑ |0.8358|± |0.0262|
105
+ | - us_foreign_policy | 0|none | 0|acc |↑ |0.8600|± |0.0349|
106
+ | - stem | 1|none | |acc |↑ |0.5623|± |0.0085|
107
+ | - abstract_algebra | 0|none | 0|acc |↑ |0.3800|± |0.0488|
108
+ | - anatomy | 0|none | 0|acc |↑ |0.6222|± |0.0419|
109
+ | - astronomy | 0|none | 0|acc |↑ |0.7039|± |0.0372|
110
+ | - college_biology | 0|none | 0|acc |↑ |0.7778|± |0.0348|
111
+ | - college_chemistry | 0|none | 0|acc |↑ |0.5400|± |0.0501|
112
+ | - college_computer_science | 0|none | 0|acc |↑ |0.5300|± |0.0502|
113
+ | - college_mathematics | 0|none | 0|acc |↑ |0.3200|± |0.0469|
114
+ | - college_physics | 0|none | 0|acc |↑ |0.4608|± |0.0496|
115
+ | - computer_security | 0|none | 0|acc |↑ |0.7800|± |0.0416|
116
+ | - conceptual_physics | 0|none | 0|acc |↑ |0.5617|± |0.0324|
117
+ | - electrical_engineering | 0|none | 0|acc |↑ |0.6138|± |0.0406|
118
+ | - elementary_mathematics | 0|none | 0|acc |↑ |0.4365|± |0.0255|
119
+ | - high_school_biology | 0|none | 0|acc |↑ |0.7839|± |0.0234|
120
+ | - high_school_chemistry | 0|none | 0|acc |↑ |0.5665|± |0.0349|
121
+ | - high_school_computer_science | 0|none | 0|acc |↑ |0.6600|± |0.0476|
122
+ | - high_school_mathematics | 0|none | 0|acc |↑ |0.4407|± |0.0303|
123
+ | - high_school_physics | 0|none | 0|acc |↑ |0.4371|± |0.0405|
124
+ | - high_school_statistics | 0|none | 0|acc |↑ |0.5602|± |0.0339|
125
+ | - machine_learning | 0|none | 0|acc |↑ |0.4643|± |0.0473|
126
+ |openbookqa | 1|none | 0|acc |↑ |0.3180|± |0.0208|
127
+ | | |none | 0|acc_norm |↑ |0.4140|± |0.0220|
128
+ |piqa | 1|none | 0|acc |↑ |0.7878|± |0.0095|
129
+ | | |none | 0|acc_norm |↑ |0.7971|± |0.0094|
130
+ |rte | 1|none | 0|acc |↑ |0.6751|± |0.0282|
131
+ |truthfulqa_mc1 | 2|none | 0|acc |↑ |0.3403|± |0.0166|
132
+ |winogrande | 1|none | 0|acc |↑ |0.7206|± |0.0126|
133
+
134
+ | Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
135
+ |------------------|------:|------|------|------|---|-----:|---|-----:|
136
+ |mmlu | 1|none | |acc |↑ |0.6411|± |0.0038|
137
+ | - humanities | 1|none | |acc |↑ |0.5896|± |0.0068|
138
+ | - other | 1|none | |acc |↑ |0.7100|± |0.0078|
139
+ | - social sciences| 1|none | |acc |↑ |0.7312|± |0.0078|
140
+ | - stem | 1|none | |acc |↑ |0.5623|± |0.0085|
141
  ```