Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / venkatasg/lil-bevo issues and pull requests

#36 - Update README.md

Pull Request - State: closed - Opened by venkatasg about 1 year ago

#35 - final updates

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#34 - updated results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#33 - Update train_clm.py

Pull Request - State: closed - Opened by juand-r over 1 year ago

#32 - updated results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#31 - Results updated with short model and ablations

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#30 - final results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#29 - updated README and spreadsheet. Deleted unnecessary files

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#28 - Update train_clm.py

Pull Request - State: closed - Opened by juand-r over 1 year ago

#27 - Larger tokenizer for 100M dataset

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#26 - Fast tokenizer with mask, cls, padding

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#25 - new Tokenizer, updated results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#25 - new Tokenizer, updated results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#24 - changed name of iter variable to match with huggingface

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#23 - updated README to be clearer

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#22 - music results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#21 - Update README.md

Pull Request - State: closed - Opened by juand-r over 1 year ago

#20 - fixed bug with abnormally low loss when running training_bevo.py

Pull Request - State: closed - Opened by juand-r over 1 year ago

#19 - updated charts and training script hyperparams

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#18 - MAESTRO results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#17 - Music pretraining scripts and results

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#16 - Added script for encoder training

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#15 - add tokenizer training wrapper script

Pull Request - State: closed - Opened by alephic over 1 year ago

#14 - Train on synthetically generated data that has some linguistic biases.

Issue - State: closed - Opened by venkatasg over 1 year ago - 1 comment

#13 - Train on audio clips of natural sounds

Issue - State: closed - Opened by venkatasg over 1 year ago - 1 comment

#12 - Bechmarked decoder models

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#11 - Interpretability of our final model

Issue - State: closed - Opened by venkatasg over 1 year ago - 1 comment

#10 - Train on music

Issue - State: closed - Opened by venkatasg over 1 year ago - 3 comments

#9 - Contrastive loss

Issue - State: closed - Opened by venkatasg over 1 year ago - 2 comments

#6 - First working version of LM

Pull Request - State: closed - Opened by venkatasg over 1 year ago

#5 - Get eval pipeline up and running

Issue - State: closed - Opened by mahowak over 1 year ago - 2 comments

#4 - Stuff we want to try for loose track

Issue - State: closed - Opened by venkatasg over 1 year ago - 1 comment

#3 - Training objectives

Issue - State: closed - Opened by venkatasg over 1 year ago

#2 - Training on non-linguistic data

Issue - State: closed - Opened by venkatasg over 1 year ago

#1 - Replicate OPT-125M BabyLM benchmark

Issue - State: closed - Opened by venkatasg over 1 year ago - 1 comment