imoneoi/openchat

Issues

• A question about the prefix of only SFT

  #211 opened 21 days ago by RikkiXu
  1

• use instructor in openchat

  #213 opened a month ago by ramandada
  1

• huggingface down: openchat can't be started

  #214 opened a month ago by antioxidanz
  1

• About performance of LLama (llama2, llama3) model

  #215 opened 23 days ago by huazhenliu
  1

• For 💻Online Demo It has been broken out.

  #207 opened 2 months ago by Schimasuperbra
  2

• opchatdataset.estimate_num_batches returns 0 at the beginning of training, and training-stuck problem

  #212 opened a month ago by syboomsy
  1

• repeat the output content until the maximum output length is set

  #209 opened 2 months ago by zestaken
  2

• Cannot create data file

  #175 opened 4 months ago by phuvinhnguyen
  1

• Does the current ochat.training_deepspeed.train only train v.3.2? How to train v.3.5?

  #173 opened 4 months ago by houghtonweihu
  1

• Single GPU vs multiple GPU (tensor parallel) suggestion for API Server

  #208 opened 2 months ago by fdm-git
  3

• Error running new model openchat-3.5-0106-gemma on RTX 4090 24GB machine, works with older mistral based model

  #198 opened 2 months ago by vikrantrathore
  2

• How was openchat/openchat-3.5-0106-gemma created?

  #197 opened 2 months ago by houghtonweihu
  2

• openchat.team is down for 3 days

  #201 opened 2 months ago by Borschik69
  1

• about data

  #206 opened 2 months ago by Luoqiu76
  0

• weighted_token_accuracy

  #205 opened 2 months ago by bino282
  0

• WARNING: Error in configuration: macro '\frac' failed its substitution!

  #204 opened 2 months ago by chengzhen123
  0

• Error when using openchat/openchat-3.5-0106-gemma in text-generation-inference

  #202 opened 2 months ago by houghtonweihu
  0

• llama_model_load: error loading model: create_tensor: tensor 'output.weight' not found

  #199 opened 2 months ago by wac81
  1

• Exception in ASGI application

  #176 opened 4 months ago by Favern
  3

• Can not Reproduce benchmarks

  #186 opened 2 months ago by zhang7346
  4

• Was the chat template applied in ochat/config/conversation_template.py?

  #200 opened 2 months ago by houghtonweihu
  0

• in llama.cpp load gguf has some issue in it

  #196 opened 2 months ago by wac81
  0

• Shall we add a Contrib Guide .md (linked and separate from README)?

  #195 opened 2 months ago by KemingHe
  0

• What is a minimum hardware requirements for setup as a sever ?

  #192 opened 2 months ago by Altenrion
  0

• How to get around the template error in OpenChat 3.5 - 0106?

  #187 opened 2 months ago by arthurfrolov
  1

• Feature request: Allow run_eval to use more than 1 GPU

  #191 opened 2 months ago by corey-lambda
  0

• How was model imone/Mistral_7B_with_EOT_token created? Any details?

  #189 opened 2 months ago by houghtonweihu
  1

• Using anything > 2048 for batch_max_length during training results in cuda index errors

  #190 opened 2 months ago by corey-lambda
  1

• Why the trained model does not produce the answer provided in the training data?

  #188 opened 2 months ago by houghtonweihu
  0

• HjsonDecodeError("Bad key name (eof)", s, end) during training

  #171 opened 4 months ago by houghtonweihu
  3

• OpenChat API Non-Deterministic Behavior

  #183 opened 2 months ago by SkanderHellal
  2

• How many epochs did you use to train openchat 3.5?

  #185 opened 3 months ago by houghtonweihu
  1

• Question for https://huggingface.co/openchat/openchat-3.5-0106/blob/main/openchat.json

  #181 opened 3 months ago by houghtonweihu
  1

• Question about `--per-sequence-loss`

  #182 opened 3 months ago by Sanster
  1

• `pip install ochat` - name 'nvcc_cuda_version' is not defined

  #164 opened 4 months ago by SantaSpeen
  4

• There is a 7B model, good at math: https://github.com/deepseek-ai/DeepSeek-Math.

  #180 opened 3 months ago by houghtonweihu
  0

• What is the meaning of padding-free in ReadMe?

  #179 opened 3 months ago by houghtonweihu
  0

• run multiple instances on a single gpu

  #178 opened 3 months ago by amir264710
  0

• I tried to pre-tokeinzation but failed

  #177 opened 3 months ago by userdsr
  0

• How to finetune openchat for downstream apps?

  #174 opened 4 months ago by houghtonweihu
  0

• Extending your training section in ReadMe?

  #163 opened 4 months ago by houghtonweihu
  0

• In the Web UI: https://openchat.team/, how can we specify "condition": "Math Correct"?

  #165 opened 4 months ago by houghtonweihu
  0

• Could the idea of Self-Play Fine-Tuning be used in C-RLFT?

  #166 opened 4 months ago by houghtonweihu
  0

• API token

  #167 opened 4 months ago by MacNoob
  1

• distill openchat

  #172 opened 4 months ago by sujitvasanth
  0

• the website requires an openAI API key to work

  #169 opened 4 months ago by HMZElidrissi
  2

• The AI on the site openchat.team is down

  #168 opened 4 months ago by Gerrytheskull
  1

• the model of Pre-tokenized dataset openchat_v3.2_super.train.parquet is Llama2 or Mistral?

  #170 opened 4 months ago by alphanlp
  1

• did ```calculate_auto_lr``` affect a lot?

  #162 opened 4 months ago by fpcsong
  4

• How do you train OpenChat on datasets GSM8K and Math?

  #161 opened 4 months ago by houghtonweihu
  0