One
imone
AI & ML interests
Reinforcement Learning, Brain-inspired AI
Professional RL(HF) Hyperparameter Tuner
Organizations
imone's activity
MMLU Lower Results Theory
3
#5 opened about 1 month ago
by
fblgit
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6401c8c9f98fbc64bcd7dca1/MOSgc_mPbfUZ-354osy1v.png)
Why is the "measured" benchmark score of Llama-3-8B so low?
1
#6 opened about 1 month ago
by
c6sneaky
MATH augmentation correctness
2
#3 opened about 2 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Answer correctness?
#11 opened about 2 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
License
8
#3 opened 2 months ago
by
mrfakename
![](https://cdn-avatars.huggingface.co/v1/production/uploads/62e54f0eae9d3f10acb95cb9/VAyk05hqB3OZWXEZW-B0q.png)
Update added_tokens.json
#8 opened 3 months ago
by
vicky4s4s
![](https://cdn-avatars.huggingface.co/v1/production/uploads/64eada6ea1e1a36f9615f37f/sinLIzrtIx8bPkbUJOFm2.jpeg)
Consider using an OSI-approved license like Mistral and Phi-2
1
#47 opened 4 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Full precision weights
6
#6 opened 5 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Which model is your demo page using?
2
#44 opened 5 months ago
by
wempoo
Freezing Issue with gguf quant
5
#1 opened 6 months ago
by
dillfrescott
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6215ce9abfcb3893344dd0a2/ez4OeVTMOpRBCZNjIufoF.jpeg)
Fix context length in config
#117 opened 6 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
MetaMath QA
1
#9 opened 6 months ago
by
mrfakename
![](https://cdn-avatars.huggingface.co/v1/production/uploads/62e54f0eae9d3f10acb95cb9/VAyk05hqB3OZWXEZW-B0q.png)
Fine Tuning
1
#8 opened 6 months ago
by
Aditya0097
Prompt template standard
1
#7 opened 6 months ago
by
Hugs4Llamas
Is there a way to get the text embedding?
1
#5 opened 6 months ago
by
EladC
What is the base model of openchat ? Llama /mistral / custom ?
4
#4 opened 6 months ago
by
StephanePop
error in docs
2
#6 opened 6 months ago
by
PsiPi
![](https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/HqUwLKO5rKA-6YilGoBwk.png)
32k context size?
1
#3 opened 6 months ago
by
paryska99
How did Mixtral make openchat_3.5 worse?
3
#34 opened 7 months ago
by
JJJJJPSYCHIC
Some feedback
1
#33 opened 7 months ago
by
cmp-nct
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6344a1b0762379fc63017e62/g4VIT8l2lZIj6AoQAwVy7.png)
🚩 Report : Ethical issue(s)
2
#1 opened about 1 year ago
by
stefan-it
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1584020801691-noauth.jpeg)
Why does this model perform so poorly on DROP compared to OpenHermes?
1
#29 opened 7 months ago
by
yahma
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1672330023435-62c6faed53c7156f5bf767ed.png)
Inconsistent Eval Results with Openchat 3.5?
2
#7 opened 7 months ago
by
banghua
Add chat template
2
#27 opened 7 months ago
by
Rocketknight1
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1660312628256-60ba519750effef3a58beac3.png)
Is this dataset generated by GPT-4?
2
#2 opened 7 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
function calling
4
#24 opened 7 months ago
by
mersahin26
![](https://cdn-avatars.huggingface.co/v1/production/uploads/647da5710ed7d0c87608d251/N0cvJlLJsBwPacTtsWJr5.jpeg)
Adding Evaluation Results
#25 opened 7 months ago
by
leaderboard-pr-bot
![](https://cdn-avatars.huggingface.co/v1/production/uploads/655506df9dc61e22c5f9c732/IZGvup0FdVlioPPIPnzZv.jpeg)
Question about openchat3.5 gsmk8 score on openllm leaderboard.
2
#23 opened 7 months ago
by
balisujohn
他这个模型有没有推理能力啊
1
#17 opened 8 months ago
by
ddls
non-commercial license
20
#1 opened 8 months ago
by
clem
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg)
Create generation_config.json
1
#21 opened 8 months ago
by
fenglui
OpenChat 3.5 few-shot results
3
#2 opened 8 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
License
15
#25 opened 8 months ago
by
mrfakename
![](https://cdn-avatars.huggingface.co/v1/production/uploads/62e54f0eae9d3f10acb95cb9/VAyk05hqB3OZWXEZW-B0q.png)
Too many zeros for GSM8K, eval prompt is not suitable for CHAT models.
13
#360 opened 8 months ago
by
JosephusCheung
What base model does it based?
2
#14 opened 8 months ago
by
lucasjin
Overfit on ChatGPT data
2
#15 opened 8 months ago
by
macadeliccc
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6455cc8d679315e4ef16fbec/M6Cfifn05BUzkCFd2QDIT.png)
Is the gsm8k evaluated few-shot (no CoT)?
2
#365 opened 8 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Why does it report an error like this when running?
2
#12 opened 8 months ago
by
Simkinhu
Update dataset details in model card
#11 opened 8 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Hallucinations
10
#2 opened 8 months ago
by
Ricepig
![](https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/_kQEH8NNG_5bahlWujd1W.png)
Great. Now make 128k version like they done with Mistral lately : )
2
#8 opened 8 months ago
by
Pumba2
Create generation_config.json
2
#9 opened 8 months ago
by
fenglui
How to setup system message
13
#5 opened 8 months ago
by
fernandofernandes
![](https://cdn-avatars.huggingface.co/v1/production/uploads/646e57a5cb6ea6e6b6df1ad4/PlGhM2SUynFBUdYAylaZK.jpeg)
EOS should be 32000
#4 opened 8 months ago
by
TheBloke
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6426d3f3a7723d62b53c259b/tvPikpAzKTKGN5wrpadOJ.jpeg)
EOS should be 32000
#3 opened 8 months ago
by
TheBloke
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6426d3f3a7723d62b53c259b/tvPikpAzKTKGN5wrpadOJ.jpeg)
This might help for your next model...
3
#6 opened 8 months ago
by
Vezora
![](https://cdn-avatars.huggingface.co/v1/production/uploads/649a54b896d5747b35e2163b/tdZmsov6fN1VHztaE5kX9.jpeg)
MMLU of ChatGPT/GPT3.5-turbo is 69~70, GSM8K 78.2
3
#1 opened 8 months ago
by
JosephusCheung
Architectural difference with Llama
1
#20 opened 9 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Dataset contamination tests
1
#1 opened 9 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Was the entire OpenOcra dataset used?
1
#9 opened 10 months ago
by
gameveloster
Difference between previous openchat
1
#1 opened 10 months ago
by
robinsongh381
System message and API model
3
#2 opened 10 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Is all of the dataset generated by gpt4, and which API version (gpt-4-0314/gpt-4-0613/gpt-4) is used?
1
#1 opened 10 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
Add type
#1 opened 10 months ago
by
osanseviero
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png)
Good model, but still struggle with riddles
4
#2 opened 11 months ago
by
gt332a
How is the coding performance?
3
#1 opened 11 months ago
by
rombodawg
![](https://cdn-avatars.huggingface.co/v1/production/uploads/642cc1c253e76b4c2286c58e/fGtQ_QeTjUgBhIT89dpUt.jpeg)
Can you explain how can we train multi-turn conversation?
3
#6 opened 11 months ago
by
tridungduong16
![](https://cdn-avatars.huggingface.co/v1/production/uploads/6454fa48b27940efcb944bb9/3GcYK4RXljPSjBUgQEArL.png)
Consider including OpenChat 3 models for human evaluation
#2 opened 11 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)
The dataset filtering script
9
#6 opened 12 months ago
by
imone
![](https://cdn-avatars.huggingface.co/v1/production/uploads/61b6cbbdbfb266841ec0f24a/PHUVNOOMEw_R2CF3u-sMS.png)