abertsch72 unlimiformer issues

abertsch72 / unlimiformer

Public repo for the NeurIPS 2023 paper "Unlimiformer: Long-Range Transformers with Unlimited Length Input"

MIT License

1.04k stars 77 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

BookSum_Full BART Baseline script/code

#66 saxenarohit opened 1 week ago
3
DatasetGenerationError

#65 pppyb closed 1 month ago
1
Unable to load dataset

#64 Ozawa333 opened 1 month ago
4
Error in running Llama 2 generation example

#63 OswaldHe opened 4 months ago
0
How can we use unlimiformer for sequence classification (textual entailment)?

#62 robinsingh-ai opened 4 months ago
0
Hardware Requirement for Running Llama-2 inferences

#61 shang-zhu opened 6 months ago
2
LLama2_example output random words

#60 KerolosAtef opened 6 months ago
1
Can't run the provided llama2 example

#59 KerolosAtef opened 6 months ago
6
GPU VRAM Usage during training

#58 KevinD777 opened 7 months ago
1
reproducing your results

#57 patrickocal opened 7 months ago
7
Prompt with Llama-2 stops after "Loading checkpoint shards: 0%"

#56 XmasRock closed 5 months ago
2
Use of other Encode/Decoder Models

#55 rdmerillat opened 8 months ago
8
IndexError when running inference with Llama-2 model

#54 shang-zhu closed 8 months ago
3
Why is the inference so slow?

#53 cckao closed 5 months ago
3
multi-gpu unlimiformer training: Expected all tensors to be on the same device

#52 shi-kejian opened 9 months ago
4
Script utilizing LLM

#51 jcgeo9 opened 9 months ago
1
Why "import sled" was commented out in run.py?

#50 shi-kejian closed 9 months ago
4
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, ....

#49 shi-kejian closed 9 months ago
0
Error Encountered While Running 'run_generation.py' Script

#48 arqumk opened 9 months ago
1
About adding a prefix and input length

#47 apapoudakis closed 5 months ago
3
Relative positions in RoPE embeddings

#46 AshwinRamachandran2002 opened 9 months ago
2
Question: During training, the calculation of topk value’s att_weight is different from the classic transformer’s multi-head attention.

#45 jjkk123456 closed 5 months ago
1
Why using different calculation methods for the key and value of the cross-attention of the decoder layer in the training and validation stages?

#44 jjkk123456 closed 10 months ago
2
Set max_size to 128 but use 512 tokens

#43 adivoj closed 10 months ago
2
error while training

#42 kekekawaii2839 closed 10 months ago
2
Errors on running llama with `test_datastore`

#41 wywyWang closed 10 months ago
8
Question:too many indices for tensor of dimension 1

#40 Lavi11C opened 10 months ago
16
API server for unlimiformer

#39 neubig opened 10 months ago
2
Running Unlimiformer with the `forward` method

#38 testzer0 opened 10 months ago
3
Fix typos

#37 szepeviktor closed 9 months ago
2
Fix changes of the training_args variable

#36 9au5a closed 10 months ago
1
Not really an issue - TrainingArguments are now immutable

#35 9au5a closed 10 months ago
2
support other llms?

#34 chaunceyliu30 closed 5 months ago
3
Steps to run the code

#33 sahulsumra opened 10 months ago
5
knn_args, unlimiformer_args, tokenizer is not defined

#32 laeljh closed 5 months ago
1
Unused variable `q_embed` in the Llama's `preprocess_query` method

#31 seunghyukoh closed 10 months ago
1
About the method `attention_forward_hook`

#30 seunghyukoh closed 11 months ago
2
running unlimiformer inference on multiple gpus

#29 kekekawaii2839 closed 5 months ago
6
Unable to produce any output with llama 2 summarization example

#28 cem2ran opened 11 months ago
1
I Will suggest you simple user interface using gradio.

#27 imrankh46 opened 11 months ago
1
Sanity check: VRAM usage on llama-2-7b-chat-hf higher than without Unlimiformer on low tokens?

#26 SharkWipf opened 11 months ago
6
TypeError: torch_replacement_knn_gpu() got an unexpected keyword argument 'device'

#25 jordancole21 opened 11 months ago
17
ImportError: cannot import name 'Unlimiformer' from 'unlimiformer'

#24 yungsinatra0 closed 11 months ago
18
Can unlimiformer work with common fine-tuning methods？

#23 mrlzh opened 1 year ago
1
Update README.md

#22 VeryG00dName closed 1 year ago
1
Encoder Only Unlimiformer

#21 YHL04 closed 11 months ago
5
Error while evaluating

#20 MonliH closed 11 months ago
2
Working with 8bit and 4bit quantized models

#19 jordancole21 opened 1 year ago
10
Support multilingual model like mt0, mBart ?

#18 trannhatquy closed 11 months ago
2
Reproduce the +test Unlimiformer setup

#17 Leonard907 closed 11 months ago
7