tiiuae/falcon-7b · Discussions

Custom 4-bit Finetuning 5-7 times faster inference than QLora

pinned

1

#13 opened over 1 year ago by

rmihaylov

How to make it work for less experienced AI whisperers

pinned

17

#4 opened over 1 year ago by

Sloba

Support for LoRA?

pinned

17

#3 opened over 1 year ago by

cekal

Fix for issue #104

#105 opened about 2 months ago by

vebM

AttributeError: 'Logger' object has no attribute 'warning_once'

#104 opened about 2 months ago by

vebM

Inference tensors cannot be saved for backward. To work around you can make a clone to get a normal tensor and use it in autograd.

#103 opened 4 months ago by

Mubarak127

Two repeated errors in model output

1

#102 opened 7 months ago by

virilo

ValueError: The current `device_map` had weights offloaded to the disk.

#101 opened 8 months ago by

MohamedZouabi

Why does falcon-7b have 71 attention heads?

1

#100 opened 8 months ago by

alpindale

Creating vectordatabse using the falcon-7b model embeddings.

#99 opened 8 months ago by

alchemistPS01

FalconForCausalLM does not support Flash Attention 2.0 yet

#98 opened 10 months ago by

Menouar

Questions

#97 opened 11 months ago by

Ppq62

Error while trying to load model

#96 opened 11 months ago by

dwojcik

Adding `safetensors` variant of this model

#95 opened 11 months ago by

SFconvertbot

Adding Evaluation Results

#94 opened 12 months ago by

leaderboard-pr-bot

Model does not know when to stop generating text?

#93 opened about 1 year ago by

jashsayani

Could we machine translatation task using this model?

2

#91 opened about 1 year ago by

Pitambarmuduli

Falcon-7B decoding error

#90 opened about 1 year ago by

rahulseetharaman

[AUTOMATED] Model Memory Requirements

#89 opened about 1 year ago by

model-sizer-bot

Upload configuration_RW.py

#88 opened about 1 year ago by

imranshah

Upload configuration_RW.py

#87 opened about 1 year ago by

imranshah

Getting: HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/tiiuae/falcon-7b/resolve/main/configuration_RW.py

1

#86 opened about 1 year ago by

f5-lolabhattu

What does this file do? modeling_falcon.py

#85 opened about 1 year ago by

Tony068

Anyone discovered "Mini" yet in prompting?

#83 opened about 1 year ago by

YoYo1234Qwerty

How to avoid running into memory/ storage problems associated with HF Spaces while using tiiuae/falcon-7b 0r 40b etc.,

4

#82 opened about 1 year ago by

vsrinivas

Update generation_config.json

1

#81 opened about 1 year ago by

nkasmanoff

ValueError: Unrecognized configuration class <class 'transformers_modules.falcon-7b.configuration_RW.RWConfig'> for this kind of AutoModel....

2

#80 opened about 1 year ago by

Inoob

RuntimeError: CUDA error: CUBLAS_STATUS_NOT_SUPPORTED

#79 opened about 1 year ago by

ConorVanek

ImportError: Using `load_in_8bit=True` requires Accelerate

3

#78 opened about 1 year ago by

aimananees

Adding `safetensors` variant of this model

#77 opened about 1 year ago by

bikalnetomi

Use input attention mask instead of casual mask in attention

#76 opened about 1 year ago by

CyberZHG

Question answering task with falcon model fails with "TypeError: forward() got an unexpected keyword argument 'token_type_ids'"

1

#75 opened about 1 year ago by

karolzak13

Inaccurate number of parameters

1

#74 opened over 1 year ago by

mohamedlotfy50

Title: Best Practice for Handling Variable-Length Sequences in Training an LLM Model on a Chatbot Dataset

#73 opened over 1 year ago by

humza-sami

Can't use the model load locally

1

#72 opened over 1 year ago by

Alouettewind

Falcon 7b instruct using cpu for inference even on NVIDIA A40 cards with 50GB VRAM

#70 opened over 1 year ago by

Akshadv

Why is alibi: false in the config.json?

#69 opened over 1 year ago by

ekurtic

getting error

2

#67 opened over 1 year ago by

Akash1267a

Revert in-library commit

#65 opened over 1 year ago by

Rocketknight1

Senior ML Scientist

#63 opened over 1 year ago by

FinTrU-TA

OSError: tiiuae/falcon-7b does not appear to have a file named configuration_RW.py

5

#62 opened over 1 year ago by

chintan4560

about eos and bos token id

1

#61 opened over 1 year ago by

louisY

configuration_RW.py missing in latest commit

9

#60 opened over 1 year ago by

ravikiran3690

Inference time issue

#59 opened over 1 year ago by

amnasher

Update generation_config.json

#55 opened over 1 year ago by

psinger

Fine-tuning issues

1

#53 opened over 1 year ago by

nebulae7

How to push or shere adapter to the hub?

7

#52 opened over 1 year ago by

Imran1

Getting an error: RuntimeError: shape '[x, 71, 64]' is invalid for input of size 3904

1

#51 opened over 1 year ago by

Carolinehu

Getting an error TypeError: unsupported operand type(s) for *: 'Tensor' and 'NoneType'

7

#49 opened over 1 year ago by

NajiAboo

Fix typo in `README.md`

#48 opened over 1 year ago by

alvarobartt