ValueError: Pipeline with tokenizer without pad_token cannot do batching. You can try to set it with `pipe.tokenizer.pad_token_id = model.config.eos_token_id`.

#39 opened 3 months ago by

jsemrau

Run this on CPU and use tool calling

#38 opened 3 months ago by

J22

!!Access Problem

#37 opened 3 months ago by

fengzi258

LLama-3.1-8B generates way to long answers!

#36 opened 3 months ago by

ayyylemao

Tokenizer error and/or 'rope_scaling' problem

#35 opened 3 months ago by

fazayjo

Deployment to Inference Endpoints

#34 opened 3 months ago by

stmackcat

Best practice for tool calling with meta-llama/Meta-Llama-3.1-8B-Instruct

#33 opened 3 months ago by

zzclynn

The model often enters infinite generation loops

#32 opened 3 months ago by

sszymczyk

unable to load 4-bit quantized varient with llama.cpp

#31 opened 3 months ago by

sunnykusawa

Garbage output ?

#30 opened 3 months ago by

danielus

Question about chat template and fine-tuning

#23 opened 3 months ago by

tblattner

Issues loading model with ooabooga textgenwebui

#20 opened 3 months ago by

Kenji776

what is the right tokenizer should I use for llama 3.1 8B?

#19 opened 3 months ago by

calebl

The sample code on the model card page is not right

#18 opened 3 months ago by

kmtao

My alternative quantizations.

#16 opened 3 months ago by

ZeroWw

ValueError: `rope_scaling` must be a dictionary with two fields

#15 opened 3 months ago by

jsemrau

Independently Benchmarked Humaneval and Evalplus scores

#13 opened 3 months ago by

VaibhavSahai

DO NOT MERGE v2 make sure vllm and transformers work

#12 opened 3 months ago by

ArthurZ

DO NOT MERGE test for vllm

#11 opened 3 months ago by

ArthurZ

Please do not include original PTH files.

#10 opened 3 months ago by

Qubitium

Utterly based

#9 opened 3 months ago by

llama-anon