mosaicml/mpt-7b · Discussions

Adding `safetensors` variant of this model

#92 opened about 1 month ago by

SFconvertbot

mpt-7b taking several minutes on mac m1?

#88 opened 4 months ago by

rmiller3

Adding `safetensors` variant of this model

#87 opened 4 months ago by

SFconvertbot

Adding Evaluation Results

#86 opened 5 months ago by

leaderboard-pr-bot

[AUTOMATED] Model Memory Requirements

#85 opened 5 months ago by

model-sizer-bot

Request: DOI

#80 opened 9 months ago by

yousefHageb

How to append new token and train?

2

#70 opened 10 months ago by

NickyNicky

fix HPU could not handle float16 in attention.py.

3

#68 opened 10 months ago by

sywangyi

After installing triton, running pipe() return "fatal error: cuda.h: No such file or directory " and "CalledProcessError: Command '['/usr/bin/gcc'...."

2

#66 opened 10 months ago by

nps798

Adding `safetensors` variant of this model

2

#65 opened 10 months ago by

makeColabFree

Converting To Flax

#64 opened 10 months ago by

erfanzar

MPT-7b on colab - RAM of GPU not used

5

#50 opened 11 months ago by

vi-c

Can support MPS device type?

1

#48 opened 11 months ago by

LouiSum

Merge cekal/mpt-7b-peft-compatible

6

#42 opened 11 months ago by

muelletm

Support gradient checkpointing

7

#41 opened 11 months ago by

muelletm

Issue training With Triton

10

#40 opened 11 months ago by

MikeyBelllissimo

Finetuning MPT-7B in 4-bit

#39 opened 11 months ago by

rmihaylov

attn_impl

11

#27 opened 11 months ago by

GaaraOtheSand

Fixes for PEFT Tuning based on iwalton3

#25 opened 11 months ago by

SebastianBodza

Can this be fine-tuned using Amazon SageMaker or run on a AMD GPU that is not CUDA-enabled?

1

#18 opened 12 months ago by

Bigshot

[Experiment] MPT 7B + LangChain Custom LLM + transformers.accelerator, on a POTATO

#16 opened 12 months ago by

saber7ooth