llama-recipes

llama-recipes copied to clipboard

Update llama_guard_version in inference.py

1

# What does this PR do? This PR updates the `llama_guard_version` argument in inference.py as it takes only string Fixes # (issue) ## Feature/Issue validation/testing Please describe the tests that...

DavidChenUIUC

Function or Tool calling

8

### 🚀 The feature, motivation and pitch Looks like Llama3 has capability to call Tools like google/bing search, Would be good to have an example script with prompt template for...

jayakumark

# What does this PR do? Fixes # (issue) ## Feature/Issue validation/testing Please describe the tests that you ran to verify your changes and relevant result summary. Provide instructions so...

statelesshz

Update xpu related device setting

2

# What does this PR do? This PR update some xpu related logic for correct support. Fixes # (issue) ## Feature/Issue validation/testing Please describe the tests that you ran to...

zhuhong61

llama3 to hf model conversion does not work

4

### System Info Hello developer, The Llama-3 model was released today. I want to convert this model to a hf model, but when I follow the readme, the following issue...

yuri-son

freezing layers have differenct behaves for different models

2

### System Info Dockerfile:  ### Information - [ ] The official example scripts - [ ] My own modified scripts ### 🐛 Describe the bug when freezing the top...

hjc3613

Added a feature that allow users to use pytorch profiler or flop_counter to measure the performance during fine-tuning.

6

# What does this PR do? Added a feature that allow users to use pytorch profiler or flop_counter to measure the performance during fine-tuning. For pytorch profiler, use --use_profiler to...

wukaixingxp

Add a README to provide links to API providers

1

# What does this PR do? Add a README.md to this directory that links to API providers that support Meta Llama. ## Before submitting - [X] This PR fixes a...

carljparker

When fine-tuning the model, the loss is greater than 1

1

### System Info pytorch2.0.1 cuda11.8 gpu 3090 ### Information - [ ] The official example scripts - [x] My own modified scripts ### 🐛 Describe the bug here's my hyperparameters...

FelexTriz

NCCL communicator error: Socket Timeout when finetuning 70B model on 2 * (8* A100(80G))

9

When fine-tuning the 70b model, I always run into an error while loading the model. Usually, after loading 4 to 10 shards (totally15 shards), the following error occurs(see Error Message)....

yguo33