Adding Evaluation Results
Browse filesThis is an automated PR created with https://huggingface.co/spaces/Weyaxi/open-llm-leaderboard-results-pr
The purpose of this PR is to add evaluation results from the Open LLM Leaderboard to your model card.
If you encounter any issues, please report them to https://huggingface.co/spaces/Weyaxi/open-llm-leaderboard-results-pr/discussions
README.md
CHANGED
|
@@ -1,22 +1,70 @@
|
|
| 1 |
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
language:
|
| 4 |
-
|
| 5 |
-
|
| 6 |
base_model: EleutherAI/pythia-31m
|
| 7 |
datasets:
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
model-index:
|
| 17 |
- name: Pythia-31M-Chat-v1
|
| 18 |
results:
|
| 19 |
-
- task:
|
| 20 |
type: text-generation
|
| 21 |
name: Text Generation
|
| 22 |
dataset:
|
|
@@ -27,13 +75,13 @@ model-index:
|
|
| 27 |
args:
|
| 28 |
num_few_shot: 25
|
| 29 |
metrics:
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
source:
|
| 34 |
-
name: Open LLM Leaderboard
|
| 35 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 36 |
-
|
|
|
|
| 37 |
type: text-generation
|
| 38 |
name: Text Generation
|
| 39 |
dataset:
|
|
@@ -43,13 +91,13 @@ model-index:
|
|
| 43 |
args:
|
| 44 |
num_few_shot: 10
|
| 45 |
metrics:
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
source:
|
| 50 |
-
name: Open LLM Leaderboard
|
| 51 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 52 |
-
|
|
|
|
| 53 |
type: text-generation
|
| 54 |
name: Text Generation
|
| 55 |
dataset:
|
|
@@ -60,13 +108,13 @@ model-index:
|
|
| 60 |
args:
|
| 61 |
num_few_shot: 5
|
| 62 |
metrics:
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
source:
|
| 67 |
-
name: Open LLM Leaderboard
|
| 68 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 69 |
-
|
|
|
|
| 70 |
type: text-generation
|
| 71 |
name: Text Generation
|
| 72 |
dataset:
|
|
@@ -77,41 +125,45 @@ model-index:
|
|
| 77 |
args:
|
| 78 |
num_few_shot: 5
|
| 79 |
metrics:
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
source:
|
|
|
|
| 84 |
name: Open LLM Leaderboard
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 86 |
-
|
| 87 |
-
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
<|im_start|>system
|
| 105 |
-
You are a highly knowledgeable assistant. Help the user as much as you can.<|im_end|>
|
| 106 |
-
<|im_start|>user
|
| 107 |
-
What are some steps I can take to become a healthier person?<|im_end|>
|
| 108 |
-
<|im_start|>assistant
|
| 109 |
-
inference:
|
| 110 |
-
parameters:
|
| 111 |
-
max_new_tokens: 250
|
| 112 |
-
penalty_alpha: 0.5
|
| 113 |
-
top_k: 2
|
| 114 |
-
repetition_penalty: 1.0016
|
| 115 |
---
|
| 116 |
|
| 117 |
# A Pythia Chat Model of 31M Parameters
|
|
@@ -227,3 +279,17 @@ DPOTrainer(
|
|
| 227 |
],
|
| 228 |
)
|
| 229 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
base_model: EleutherAI/pythia-31m
|
| 6 |
datasets:
|
| 7 |
+
- totally-not-an-llm/EverythingLM-data-V3
|
| 8 |
+
- databricks/databricks-dolly-15k
|
| 9 |
+
- THUDM/webglm-qa
|
| 10 |
+
- starfishmedical/webGPT_x_dolly
|
| 11 |
+
- Amod/mental_health_counseling_conversations
|
| 12 |
+
- sablo/oasst2_curated
|
| 13 |
+
- cognitivecomputations/wizard_vicuna_70k_unfiltered
|
| 14 |
+
- mlabonne/chatml_dpo_pairs
|
| 15 |
+
pipeline_tag: text-generation
|
| 16 |
+
widget:
|
| 17 |
+
- text: '<|im_start|>system
|
| 18 |
+
|
| 19 |
+
You are a career counselor. The user will provide you with an individual looking
|
| 20 |
+
for guidance in their professional life, and your task is to assist them in determining
|
| 21 |
+
what careers they are most suited for based on their skills, interests, and experience.
|
| 22 |
+
You should also conduct research into the various options available, explain the
|
| 23 |
+
job market trends in different industries, and advice on which qualifications
|
| 24 |
+
would be beneficial for pursuing particular fields.<|im_end|>
|
| 25 |
+
|
| 26 |
+
<|im_start|>user
|
| 27 |
+
|
| 28 |
+
Heya!<|im_end|>
|
| 29 |
+
|
| 30 |
+
<|im_start|>assistant
|
| 31 |
+
|
| 32 |
+
Hi! How may I help you?<|im_end|>
|
| 33 |
+
|
| 34 |
+
<|im_start|>user
|
| 35 |
+
|
| 36 |
+
I am interested in developing a career in software engineering. What would you
|
| 37 |
+
recommend me to do?<|im_end|>
|
| 38 |
+
|
| 39 |
+
<|im_start|>assistant'
|
| 40 |
+
- text: '<|im_start|>system
|
| 41 |
+
|
| 42 |
+
You are a helpful assistant who answers user''s questions with details and curiosity.<|im_end|>
|
| 43 |
+
|
| 44 |
+
<|im_start|>user
|
| 45 |
+
|
| 46 |
+
What are some potential applications for quantum computing?<|im_end|>
|
| 47 |
+
|
| 48 |
+
<|im_start|>assistant'
|
| 49 |
+
- text: '<|im_start|>system
|
| 50 |
+
|
| 51 |
+
You are a highly knowledgeable assistant. Help the user as much as you can.<|im_end|>
|
| 52 |
+
|
| 53 |
+
<|im_start|>user
|
| 54 |
+
|
| 55 |
+
What are some steps I can take to become a healthier person?<|im_end|>
|
| 56 |
+
|
| 57 |
+
<|im_start|>assistant'
|
| 58 |
+
inference:
|
| 59 |
+
parameters:
|
| 60 |
+
max_new_tokens: 250
|
| 61 |
+
penalty_alpha: 0.5
|
| 62 |
+
top_k: 2
|
| 63 |
+
repetition_penalty: 1.0016
|
| 64 |
model-index:
|
| 65 |
- name: Pythia-31M-Chat-v1
|
| 66 |
results:
|
| 67 |
+
- task:
|
| 68 |
type: text-generation
|
| 69 |
name: Text Generation
|
| 70 |
dataset:
|
|
|
|
| 75 |
args:
|
| 76 |
num_few_shot: 25
|
| 77 |
metrics:
|
| 78 |
+
- type: acc_norm
|
| 79 |
+
value: 22.7
|
| 80 |
+
name: normalized accuracy
|
| 81 |
source:
|
|
|
|
| 82 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 83 |
+
name: Open LLM Leaderboard
|
| 84 |
+
- task:
|
| 85 |
type: text-generation
|
| 86 |
name: Text Generation
|
| 87 |
dataset:
|
|
|
|
| 91 |
args:
|
| 92 |
num_few_shot: 10
|
| 93 |
metrics:
|
| 94 |
+
- type: acc_norm
|
| 95 |
+
value: 25.6
|
| 96 |
+
name: normalized accuracy
|
| 97 |
source:
|
|
|
|
| 98 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 99 |
+
name: Open LLM Leaderboard
|
| 100 |
+
- task:
|
| 101 |
type: text-generation
|
| 102 |
name: Text Generation
|
| 103 |
dataset:
|
|
|
|
| 108 |
args:
|
| 109 |
num_few_shot: 5
|
| 110 |
metrics:
|
| 111 |
+
- type: acc
|
| 112 |
+
value: 23.24
|
| 113 |
+
name: accuracy
|
| 114 |
source:
|
|
|
|
| 115 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 116 |
+
name: Open LLM Leaderboard
|
| 117 |
+
- task:
|
| 118 |
type: text-generation
|
| 119 |
name: Text Generation
|
| 120 |
dataset:
|
|
|
|
| 125 |
args:
|
| 126 |
num_few_shot: 5
|
| 127 |
metrics:
|
| 128 |
+
- type: acc
|
| 129 |
+
value: 47.99
|
| 130 |
+
name: accuracy
|
| 131 |
source:
|
| 132 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 133 |
name: Open LLM Leaderboard
|
| 134 |
+
- task:
|
| 135 |
+
type: text-generation
|
| 136 |
+
name: Text Generation
|
| 137 |
+
dataset:
|
| 138 |
+
name: TruthfulQA (0-shot)
|
| 139 |
+
type: truthful_qa
|
| 140 |
+
config: multiple_choice
|
| 141 |
+
split: validation
|
| 142 |
+
args:
|
| 143 |
+
num_few_shot: 0
|
| 144 |
+
metrics:
|
| 145 |
+
- type: mc2
|
| 146 |
+
value: 0.0
|
| 147 |
+
source:
|
| 148 |
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 149 |
+
name: Open LLM Leaderboard
|
| 150 |
+
- task:
|
| 151 |
+
type: text-generation
|
| 152 |
+
name: Text Generation
|
| 153 |
+
dataset:
|
| 154 |
+
name: GSM8k (5-shot)
|
| 155 |
+
type: gsm8k
|
| 156 |
+
config: main
|
| 157 |
+
split: test
|
| 158 |
+
args:
|
| 159 |
+
num_few_shot: 5
|
| 160 |
+
metrics:
|
| 161 |
+
- type: acc
|
| 162 |
+
value: 0.0
|
| 163 |
+
name: accuracy
|
| 164 |
+
source:
|
| 165 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Felladrin/Pythia-31M-Chat-v1
|
| 166 |
+
name: Open LLM Leaderboard
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
---
|
| 168 |
|
| 169 |
# A Pythia Chat Model of 31M Parameters
|
|
|
|
| 279 |
],
|
| 280 |
)
|
| 281 |
```
|
| 282 |
+
|
| 283 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 284 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Felladrin__Pythia-31M-Chat-v1)
|
| 285 |
+
|
| 286 |
+
| Metric |Value|
|
| 287 |
+
|---------------------------------|----:|
|
| 288 |
+
|Avg. |19.92|
|
| 289 |
+
|AI2 Reasoning Challenge (25-Shot)|22.70|
|
| 290 |
+
|HellaSwag (10-Shot) |25.60|
|
| 291 |
+
|MMLU (5-Shot) |23.24|
|
| 292 |
+
|TruthfulQA (0-shot) | 0.00|
|
| 293 |
+
|Winogrande (5-shot) |47.99|
|
| 294 |
+
|GSM8k (5-shot) | 0.00|
|
| 295 |
+
|