nvidia/NV-Embed-v1 · Discussions

Can't load model with SentenceTransformers 3.0.1 AttributeError: 'LatentAttentionConfig' object has no attribute '_attn_implementation_internal'

8

#50 opened about 1 month ago by

jswarner85

Can we get a simple example please using "huggingfaceembeddings" from the Langchain library or whatever class is the correct one?

#48 opened about 2 months ago by

ctranslate2-4you

Update feature for NVEmbedConfig class

1

#45 opened 2 months ago by

lukelv

Batch_size

#44 opened 2 months ago by

lukelv

replicate experimental results on the MTEB dataset

1

#42 opened 2 months ago by

lzq2021

Code trying to download model from huggingface instead of using Locally Downloaded Model

4

#41 opened 3 months ago by

sharedJackpot

Model Loading Error

3

#40 opened 3 months ago by

kcsham

Supporting Flash Attention 2.0

#39 opened 3 months ago by

Cdemir

'MistralModel' object has no attribute 'encode'

1

#38 opened 3 months ago by

dadada

Has the tokenizer of the base model(Mistral-7B-v0.1) been retrained?

#37 opened 3 months ago by

LH0521

How did you trained your LatentAttentionLayer?

1

#36 opened 3 months ago by

juneonetwothree

Why do we need to hardcode self._attn_implementation = "eager"

1

#35 opened 3 months ago by

shantanuagarwal

Error to load model with HuggingFace API

1

#34 opened 3 months ago by deleted

Regarding max seq length

1

#33 opened 3 months ago by

sandeep456

How to fine-tune this model?

#32 opened 3 months ago by

caochengchen

error with module datasets

2

#31 opened 3 months ago by

claraadam

Distant resource does not have a Content-Length

#30 opened 3 months ago by

caochengchen

Best instructions for clustering and semantic similarity

2

#29 opened 3 months ago by

rmilliere

Dataloader multiprocessing error

1

#28 opened 3 months ago by

Atsunori

Fixing "KeyError: 'NVEmbedConfig'"

9

#27 opened 3 months ago by

Th3l

Error using multi-gpu support

4

#26 opened 4 months ago by

bobwhiterabbit

Access to model nvidia/NV-Embed-v1 is restricted. You must be authenticated to access it

6

#25 opened 4 months ago by

yijiu

Matryoshka Embedding

1

#24 opened 4 months ago by

XingyanZhang

nvidia/NV-Embed-v1 is not the path to a directory containing a file named config.json.

3

#23 opened 4 months ago by

XuehangCang

Finetuning guidelines

#21 opened 4 months ago by

mali404

How much VRAM is needed to run this model? Like for the bare minimum length etc?

3

#20 opened 4 months ago by

smpa239

Ollama Version

1

#19 opened 4 months ago by

yangwang825

Weights are in FP16 (loaded in FP32) but paper mentions BF16

#17 opened 4 months ago by

AdrienC

ONNX version

1

#16 opened 4 months ago by

michaelfeil

Sentence Transformer compatibility

4

#15 opened 4 months ago by

michaelfeil

Please provide a 8bit quantified version

#14 opened 4 months ago by

fukai

How to use for AutoModelForSequenceClassification?

#13 opened 4 months ago by

deshwalmahesh

Possible to implement `_no_split_modules` attribute?

1

#12 opened 4 months ago by

ronnybehrens

missing citation

3

#11 opened 4 months ago by

SeanLee97

Multi-Lingual?

2

#10 opened 4 months ago by

dejanseo

Getting "KeyError" when loading model

5

#8 opened 4 months ago by

tsakaiba

TypeError: MistralDecoderLayer.forward() got an unexpected keyword argument 'is_causal'

3

#7 opened 4 months ago by

yxzwayne

Is this model active?

1

#5 opened 4 months ago by

gsnic

Sharing training data & reproducing training

1

#4 opened 4 months ago by

xhluca