databrickslabs/dolly

Issues

• Can dolly support torch.jit.trace?

  #213 opened 8 months ago by JackWeiw
  0

• step by step instructions on how to build this AI from source code

  #212 opened a year ago by adrelanos
  0

• Tensor sizes are printed while running .generate function

  #211 opened a year ago by chopra-anuj
  0

• Changing the random seed of the dolly model

  #210 opened a year ago by promiseve
  1

• deploy dolly langchain demo to mlflow issue

  #208 opened a year ago by nigelhussain
  2

• Deepspeed version is not supported anymore

  #209 opened a year ago by ivnbarrera-ctl
  2

• Out of Memory in A100 Google Colab Pro+

  #207 opened a year ago by jamesjohnson1025
  3

• Please clarify the License Policy

  #206 opened a year ago by JayLiangs
  1

• Training dolly in A100 (Google colab)

  #205 opened a year ago by jamesjohnson1025
  1

• LLM to analyze DataFrame

  #204 opened a year ago by seshsan
  1

• Nvidia H100 - Fine tune Dolly Failure - "Loss" and "learning rate" are start at 0 and remain 0 continually

  #202 opened a year ago by GeorgiAngelov
  1

• Does not work on Windows 11

  #203 opened a year ago by dustin2551
  1

• How can I cite your work?

  #201 opened a year ago by Patrick-Ni
  1

• HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': 's3://skillsai-llms/dolly_LLM/dolly-7b-cust_trained_v5'. Use `repo_type` argument if needed.

  #198 opened 2 years ago by deepakkrish91
  1

• Facing error while training on custom data using deepspeed

  #199 opened 2 years ago by kapilmayank
  2

• how to train the model locally

  #196 opened 2 years ago by SeekPoint
  1

• Matching the response tokens

  #197 opened 2 years ago by vwxyzjn
  1

• load_in_8bit

  #195 opened 2 years ago by d3ath-add3r
  2

• GPU g5.4xlarge, training pythia-2.8b model exits with return code = -9

  #194 opened 2 years ago by ipfloater
  2

• what should i change when i want to use my own datasets

  #190 opened 2 years ago by AndyLinOuO
  2

• Dolly Model Register on Mflow and Serve using Databricks Model Serving

  #191 opened 2 years ago by SrikarPrayaga06
  1

• [Question][Proposal] Libgreattao integration

  #188 opened 2 years ago by nintyfan
  4

• FValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name':

  #187 opened 2 years ago by ryurobin1990
  1

• How to use domain specific data?

  #185 opened 2 years ago by khayamgondal
  3

• 'GPTNeoXForCausalLM' object has no attribute 'generation_config'

  #184 opened 2 years ago by westonli-thu
  1

• hi！how to cite this work？

  #183 opened 2 years ago by xuyifan-0731
  1

• generate_response with context

  #177 opened 2 years ago by ebilal
  1

• databricks demos with dolly NOT working

  #178 opened 2 years ago by Free-Radical
  2

• Training on 500k data

  #179 opened 2 years ago by jawadSajid
  1

• General Rule of Thumb when Estimating GPU Memory Requirements

  #176 opened 2 years ago by Hegelim
  4

• Using dolly-v2-7b and generating 5000k words I get this error: The size of tensor a (2048) must match the size of tensor b (2049) at non-singleton dimension 3

  #174 opened 2 years ago by zubair-ahmed-ai
  12

• OSError: does not appear to have a file named config.json.

  #175 opened 2 years ago by sp496
  1

• How long does it take for training 2 epochs on 8 x NVIDIA A100

  #171 opened 2 years ago by boogiegames3
  10

• [HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name']

  #173 opened 2 years ago by Umesh1307
  1

• OSError: does not appear to have a file named config.json. Checkout

  #167 opened 2 years ago by Umesh1307
  2

• Is this correct approach to do Prompt Tuning on DollyV2 model

  #158 opened 2 years ago by pratikchhapolika
  12

• training: CUDA out of memory

  #170 opened 2 years ago by vincentvic
  2

• TypeError: __init__() got an unexpected keyword argument 'load_in_8bit'

  #168 opened 2 years ago by chintan-donda
  15

• Clarification on GPU Accelerated compute

  #172 opened 2 years ago by dineshreddy221
  1

• RuntimeError: Error building extension 'fused_adam'

  #156 opened 2 years ago by chintan-donda
  7

• Getting error when running deepseed in dolly training with exits return code = -9

  #166 opened 2 years ago by dineshreddy221
  4

• How to run model and finetune it.

  #169 opened 2 years ago by nageshpindi
  1

• Invoking deepspeed with HF token converts flags to literal values

  #165 opened 2 years ago by opyate
  12

• JSONDecodeError on running dolly repo with review dataset.

  #161 opened 2 years ago by dineshreddy221
  2

• Cannot Load Dataset

  #160 opened 2 years ago by jawadSajid
  2

• Training on A100

  #152 opened 2 years ago by Louis-y-nlp
  9

• is train_dolly.py inference only?

  #154 opened 2 years ago by julianfaraone
  5

• Can I take a large corpus of text not necessary in the form of instructions and response, and just fine tune Dolly?

  #155 opened 2 years ago by abhisheksgumadi
  7

• Running deepspeed in subdirectory

  #153 opened 2 years ago by BrandenOlson
  4

• TypeError:pipeline() missing 1 required positional argument :'task'

  #151 opened 2 years ago by Minjuner-97
  1