Pokai Chang LLaMA-LoRA-Tuner Issues

zetavg/LLaMA-LoRA-Tuner Issues

TypeError: __init__() got an unexpected keyword argument 'llm_int8_skip_modules'
Updated 6 months ago
Private Repo can't load
Updated 6 months ago2
Run pretrained model
Updated 8 months ago
LlamaTokenizer class issue
Updated 10 months ago
WandB charts lose data.
Updated 10 months ago
WandB logging doesn't save evaluation loss
Updated 10 months ago
Loss chart errors after a bit
Updated 10 months ago
Anyone get this to install / run on anywhere but Colab?
Updated a year ago1
offload-between-cpu-and-gpu
Updated a year ago
Does not work any longer on Google Colab or locally
Updated a year ago1
Error when fine tuning: Size Mismatch
Updated a year ago
Executing tuned Model locally
Updated a year ago
QLora support
Updated a year ago1
Falcon Support
Updated a year ago
I always running into those error messages when installing dependencies.
Updated a year ago1
Support for Larger Models
Updated a year ago2
Use finetuned model for inference programmatically
Closed a year ago10
Continue from checkpoint fails
Updated a year ago
Running locally without git and intternet
Updated a year ago3
Can't use secondary Google drive
Updated a year ago3
Comparison to text-generation-webui
Closed a year ago3
Inference got error "probability tensor contains either `inf`, `nan` or element < 0" while using beams=2 with temperature=0.1
Updated a year ago2
Aborting training should delete output folder
Updated a year ago3
Cannot select `None` template
Updated a year ago2
Refresh instead of timing out
Updated a year ago3
Support smaller models
Closed a year ago2
Syntax error in code from local install
Closed a year ago3
Type Issue on Google Colab
Closed a year ago1
App starts but shows Error on both LoRa Model and Prompt Template fields
Closed a year ago2
Text looping when using inference and go full determinism
Closed a year ago5
ValueError('Loading ... requires you to execute the tokenizer file in that repo on your local machine. Make sure you have read the code there to avoid malicious use, then set the option `trust_remote_code=True` to remove this error.')
Closed a year ago1
Error When Fine Tuning
Closed a year ago3

zetavg / LLaMA-LoRA-Tuner

zetavg/LLaMA-LoRA-Tuner Issues

TypeError: init() got an unexpected keyword argument 'llm_int8_skip_modules'

Private Repo can't load

Run pretrained model

LlamaTokenizer class issue

WandB charts lose data.

WandB logging doesn't save evaluation loss

Loss chart errors after a bit

Anyone get this to install / run on anywhere but Colab?

offload-between-cpu-and-gpu

Does not work any longer on Google Colab or locally

Error when fine tuning: Size Mismatch

Executing tuned Model locally

QLora support

Falcon Support

I always running into those error messages when installing dependencies.

Support for Larger Models

Use finetuned model for inference programmatically

Continue from checkpoint fails

Running locally without git and intternet

Can't use secondary Google drive

Comparison to text-generation-webui

Inference got error "probability tensor contains either `inf`, `nan` or element < 0" while using beams=2 with temperature=0.1

Aborting training should delete output folder

Cannot select `None` template

Refresh instead of timing out

Support smaller models

Syntax error in code from local install

Type Issue on Google Colab

App starts but shows Error on both LoRa Model and Prompt Template fields

Text looping when using inference and go full determinism

ValueError('Loading ... requires you to execute the tokenizer file in that repo on your local machine. Make sure you have read the code there to avoid malicious use, then set the option `trust_remote_code=True` to remove this error.')

Error When Fine Tuning