Hugging Face's logo

MaziyarPanahi
/
calme-2.3-llama3-70b

Text Generation
Transformers
Safetensors
PyTorch
English
llama
axolotl
finetune
dpo
facebook
meta
llama-3
chatml
conversational
Eval Results
text-generation-inference
Model card Files Community
10
MaziyarPanahi commited on
Commit
a1b2e5d
•
1 Parent(s): ac4215c

Update README.md (#8)

Browse files

- Update README.md (3da57c1254c5274dc6267a12543431a8fc3c9842)

Files changed (1) hide show
  1. README.md +15 -9
README.md CHANGED
@@ -15,7 +15,7 @@ tags:
15
  - chatml
16
  base_model: meta-llama/Meta-Llama-3-70B-Instruct
17
  datasets:
18
- - Intel/orca_dpo_pairs
19
  model_name: Llama-3-70B-Instruct-DPO-v0.3
20
  pipeline_tag: text-generation
21
  license_name: llama3
@@ -41,7 +41,8 @@ model-index:
41
  value: 72.35
42
  name: normalized accuracy
43
  source:
44
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
45
  name: Open LLM Leaderboard
46
  - task:
47
  type: text-generation
@@ -54,10 +55,11 @@ model-index:
54
  num_few_shot: 10
55
  metrics:
56
  - type: acc_norm
57
- value: 86.0
58
  name: normalized accuracy
59
  source:
60
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
61
  name: Open LLM Leaderboard
62
  - task:
63
  type: text-generation
@@ -74,7 +76,8 @@ model-index:
74
  value: 80.47
75
  name: accuracy
76
  source:
77
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
78
  name: Open LLM Leaderboard
79
  - task:
80
  type: text-generation
@@ -90,7 +93,8 @@ model-index:
90
  - type: mc2
91
  value: 63.45
92
  source:
93
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
94
  name: Open LLM Leaderboard
95
  - task:
96
  type: text-generation
@@ -107,7 +111,8 @@ model-index:
107
  value: 82.95
108
  name: accuracy
109
  source:
110
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
111
  name: Open LLM Leaderboard
112
  - task:
113
  type: text-generation
@@ -124,7 +129,8 @@ model-index:
124
  value: 87.19
125
  name: accuracy
126
  source:
127
- url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
 
128
  name: Open LLM Leaderboard
129
  ---
130
 
@@ -233,4 +239,4 @@ outputs = pipeline(
233
  top_p=0.95,
234
  )
235
  print(outputs[0]["generated_text"][len(prompt):])
236
- ```
 
15
  - chatml
16
  base_model: meta-llama/Meta-Llama-3-70B-Instruct
17
  datasets:
18
+ - MaziyarPanahi/truthy-dpo-v0.1-axolotl
19
  model_name: Llama-3-70B-Instruct-DPO-v0.3
20
  pipeline_tag: text-generation
21
  license_name: llama3
 
41
  value: 72.35
42
  name: normalized accuracy
43
  source:
44
+ url: >-
45
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
46
  name: Open LLM Leaderboard
47
  - task:
48
  type: text-generation
 
55
  num_few_shot: 10
56
  metrics:
57
  - type: acc_norm
58
+ value: 86
59
  name: normalized accuracy
60
  source:
61
+ url: >-
62
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
63
  name: Open LLM Leaderboard
64
  - task:
65
  type: text-generation
 
76
  value: 80.47
77
  name: accuracy
78
  source:
79
+ url: >-
80
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
81
  name: Open LLM Leaderboard
82
  - task:
83
  type: text-generation
 
93
  - type: mc2
94
  value: 63.45
95
  source:
96
+ url: >-
97
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
98
  name: Open LLM Leaderboard
99
  - task:
100
  type: text-generation
 
111
  value: 82.95
112
  name: accuracy
113
  source:
114
+ url: >-
115
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
116
  name: Open LLM Leaderboard
117
  - task:
118
  type: text-generation
 
129
  value: 87.19
130
  name: accuracy
131
  source:
132
+ url: >-
133
+ https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.3
134
  name: Open LLM Leaderboard
135
  ---
136
 
 
239
  top_p=0.95,
240
  )
241
  print(outputs[0]["generated_text"][len(prompt):])
242
+ ```