Performance Discrepancy Between Local Llama 3 Deployment and Hugging Face Demo

🔥➕ 2

#152 opened almost 2 years ago by

depsemt

Is there an Updated version of LLaMa-3-8B-Instruct in the works?

#150 opened almost 2 years ago by

Joseph717171

Llama-3-Instruct with Langchain keeps talking to itself

➕👍 5

#147 opened almost 2 years ago by

fahim9778

About llama3 tokenizer

#146 opened almost 2 years ago by

Yingshu

Need urgent help with connecting with model, was working previously

#145 opened almost 2 years ago by

amit15gupta

Will there be an updated version of Llama-3-8B-Instruct?

#144 opened almost 2 years ago by

Joseph717171

Cannot load tokenizer： Exception: data did not match any variant of untagged enum PostProcessorWrapper at line 2395 column 3

#143 opened almost 2 years ago by

huangleiabcde

Fine tuned Llama3 doesn't stop generating tokens when it should.

#142 opened almost 2 years ago by

keyuisai

Error while deserializing header: HeaderTooLarge

#141 opened almost 2 years ago by

kdavid109

Using Llama offline with transformers is slow and response is repeating itself

#139 opened almost 2 years ago by

AyoxRay

Request to access the repo

#138 opened almost 2 years ago by

anatris

Meta-Llama-3-8B-Instruct: "max_new_tokens" is not working for /v1/chat/completions

🔥 1

#137 opened almost 2 years ago by

kuswe

Add co2_eq_emissions to README so that the HF API exposes this info to users.

#136 opened almost 2 years ago by

selenecodes

Despite latest transformers version '4.41.1', I get the error: The model 'LlamaModel' is not supported for text-generation

#134 opened almost 2 years ago by

guptaansit

Changing maximum input length

#133 opened almost 2 years ago by

shipWr3ck

How to introduce stop_strings in llama3?

#132 opened almost 2 years ago by

Srinjoy

*** OSError: meta-llama/Meta-Llama-3-8B-Instruct does not appear to have a file named pytorch_model.bin, tf_model.h5, model.ckpt or flax_model.msgpack.

#131 opened almost 2 years ago by

akjagadish

use inference api

#127 opened almost 2 years ago by

Miralaeeop1

How to stop Llama from generating follow-up questions and excessive conversation with langchain

#125 opened almost 2 years ago by

pelumifagbemi

Compatibility with llama-cpp-python and Ollama

#124 opened almost 2 years ago by

liashchynskyi

Cloning 'Meta-Llama-3-8B-Instruct' taking excessively long time

#121 opened almost 2 years ago by

guptaansit

Discrepancy between Base and Instruct model eos_token.

👀 1

#119 opened almost 2 years ago by

richardlian

System prompt during Fine Tuning

#117 opened almost 2 years ago by

matteoperiani

Local configuration with LlamaCpp

#116 opened almost 2 years ago by

alessandervs

login does not work on fine tuned model

🔥 1

#114 opened almost 2 years ago by

Dave-theGr8

Original Directory

#113 opened almost 2 years ago by

raunakkumar

It can be used with JavaScript if yes, how can I assign a role and some custom parameters

#111 opened almost 2 years ago by

vdggds

Request: DOI

#110 opened almost 2 years ago by

SushantGautam

tokenizer.model can't be loaded by SentencePiece: "RuntimeError: Internal: could not parse ModelProto from tokenizer.model"

👍 6

#109 opened almost 2 years ago by

ericx134

PrivateGPT settings

#108 opened almost 2 years ago by

SelfHostHub

Possible stopping issues?

#106 opened almost 2 years ago by

Dampfinchen

Tokenizer Chat Template

#103 opened almost 2 years ago by

Sm1Ling

It`s a great job ~!!!!!!

#102 opened almost 2 years ago by

MallemJerry

Why there is not pad token?

#101 opened almost 2 years ago by

Imran1

Slow API Calls in Transformers

🤝 3

#98 opened almost 2 years ago by

Zifeng01code

Llama3 8b related model has bursting issues when it generates emotion-related words

👍 2

#95 opened almost 2 years ago by

HyperBlaze

Where can I get a config.json file for Meta-Llama-3-8B-instruct?

#93 opened almost 2 years ago by

gdb94086

Local Mobile Device Hosting

#91 opened almost 2 years ago by

suguspnk

How should names & fewshot examples be encoded?

👀 8

#89 opened almost 2 years ago by

theobjectivedad

Please provide some guidance or documentation on tool usage.

👍 2

#88 opened almost 2 years ago by

lcahill

AttributeError: type object 'AttentionMaskConverter' has no attribute '_ignore_causal_mask_sdpa'

#87 opened almost 2 years ago by

gy19

Your request to access this repo has been successfully submitted, and is pending a review from the repo's authors.

#86 opened almost 2 years ago by

g3stz

Model Generating Prefix

👍 1

#85 opened almost 2 years ago by

dongyulin

cant get access

#84 opened almost 2 years ago by

ERmak158

Meta-Llama-3-8B-Instruct does not appear to have a file named config.json

#82 opened almost 2 years ago by

RK-RK

Access issues via Meta website and HF

#81 opened almost 2 years ago by

GiorgioDiSalvo

Plans for a 13B model?

#80 opened almost 2 years ago by

Cheeto96

not getting the right output

#79 opened almost 2 years ago by

Sardarmoazzam

what memory needed ?

#78 opened almost 2 years ago by

francescofiamingo

Key Error 'llma' in configuration_auto.py

#76 opened almost 2 years ago by

himasrikode