Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / princeton-nlp/autocompressors issues and pull requests
#29 - The usage of past_key_values in AutoCompressorMixin
Issue -
State: open - Opened by RewindL about 1 month ago
- 1 comment
#29 - The usage of past_key_values in AutoCompressorMixin
Issue -
State: open - Opened by RewindL about 1 month ago
- 1 comment
#28 - Does Auto-Compressors support newer base model like LLama-3 or Qwen-2.5?
Issue -
State: closed - Opened by RewindL about 2 months ago
- 2 comments
#28 - Does Auto-Compressors support newer base model like LLama-3 or Qwen-2.5?
Issue -
State: closed - Opened by RewindL about 2 months ago
- 2 comments
#27 - Evaluation datasets needed
Issue -
State: open - Opened by Mr-lonely0 2 months ago
#26 - fix: typo
Pull Request -
State: closed - Opened by khs0415p 2 months ago
#25 - Your shared model trained on LLAMA2 is not trained on Lora, It's full-finetuned model.
Issue -
State: closed - Opened by jason9693 3 months ago
- 1 comment
#24 - torchrun error when generating training split
Issue -
State: open - Opened by OswaldHe 4 months ago
- 3 comments
#23 - substep & segment
Issue -
State: closed - Opened by Lu-kuan-lpk 4 months ago
- 1 comment
#22 - Install as python package?
Issue -
State: closed - Opened by creisle 5 months ago
- 1 comment
#21 - Some issue about ICL Experience
Issue -
State: closed - Opened by broalantaps 7 months ago
- 3 comments
#20 - Inquire on data of Table 1
Issue -
State: open - Opened by void-b583x2-NULL 8 months ago
- 1 comment
#19 - Question about the data preprocessing
Issue -
State: closed - Opened by hxs91 8 months ago
- 1 comment
#18 - Reduce the number of summary vectors
Issue -
State: closed - Opened by rahulseetharaman 8 months ago
- 2 comments
#17 - question about `position_ids`
Issue -
State: closed - Opened by hxs91 8 months ago
- 5 comments
#16 - Held-out perplexity question
Issue -
State: closed - Opened by broalantaps 8 months ago
- 3 comments
#15 - RuntimeError: FlashAttention only support fp16 and bf16 data type
Issue -
State: closed - Opened by stdKonjac 10 months ago
- 3 comments
#14 - Dimension of last_hidden_state size
Issue -
State: closed - Opened by imbalu007 10 months ago
- 2 comments
#13 - AttributeError: 'SubstepTrainer' object has no attribute 'do_grad_scaling'
Issue -
State: closed - Opened by msclar 10 months ago
- 3 comments
#12 - Install instructions are not clear
Issue -
State: closed - Opened by imbalu007 10 months ago
- 2 comments
#11 - Finetuning an autocompressor model
Issue -
State: closed - Opened by imbalu007 10 months ago
- 4 comments
#10 - Some fixes to make Llama train after the merge
Pull Request -
State: closed - Opened by mu-arkhipov 11 months ago
- 1 comment
#9 - Merge Llama branch
Pull Request -
State: closed - Opened by CodeCreator 11 months ago
#8 - BUG REPORT
Issue -
State: closed - Opened by Patrick-Ni about 1 year ago
- 1 comment
#7 - Summary Vector Failures and Incomplete Answers with Numerical Contexts
Issue -
State: closed - Opened by iseesaw over 1 year ago
- 4 comments
#6 - CUDA out of memory.
Issue -
State: closed - Opened by xuguohai over 1 year ago
- 3 comments
#5 - Question on the preprocessed data
Issue -
State: closed - Opened by LouChao98 over 1 year ago
- 3 comments
#4 - Inquiry for the release date of the pre-trained model
Issue -
State: closed - Opened by siyuhsu over 1 year ago
- 1 comment
#3 - Follow-up on Code Release Timeline
Issue -
State: closed - Opened by mpoemsl over 1 year ago
- 4 comments
#2 - Timeline for Release of Code and Pre-Trained Models
Issue -
State: closed - Opened by mpoemsl over 1 year ago
- 2 comments
#1 - Readme: Add link and abstract of paper
Pull Request -
State: closed - Opened by EwoutH over 1 year ago
- 2 comments