Ecosyste.ms: Issues

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

GitHub / ayaka14732/llama-2-jax issues and pull requests

#28 - Llama2 fine-tuned chat version

Issue - State: open - Opened by ksmdnl 8 months ago

#27 - Training From Scrach -- Redpajama

Issue - State: open - Opened by opooladz 11 months ago

#26 - Unable to shard llama 13B (and 70B) on v4-32 TPU

Issue - State: open - Opened by defdet 11 months ago

#25 - Allow for transfer learning

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

#24 - Update

Pull Request - State: closed - Opened by divyapatel4 about 1 year ago

#23 - correct the formula for k

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

#22 - Problems sharding Llama-70B on TPU v3-32

Issue - State: open - Opened by divyapatel4 about 1 year ago - 1 comment

#20 - DEADLINE_EXCEEDED when running train.py on GPU node

Issue - State: open - Opened by zigzagcai about 1 year ago

#18 - Improve generation speed and add benchmark for generation

Pull Request - State: open - Opened by ayaka14732 over 1 year ago

#17 - Fix training

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#16 - forward_llama() missing 1 required keyword-only argument: 'rotary_values'

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

#15 - Generation speed

Issue - State: open - Opened by sh0416 over 1 year ago - 10 comments

#14 - Got jax.errors.TracerIntegerConversionError when running generate.py

Issue - State: open - Opened by zhangzx-uiuc over 1 year ago - 2 comments

#13 - Implement left padding

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#12 - Implement KV cache

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#11 - Update

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#10 - train.py OOM on TPUv3-8

Issue - State: open - Opened by ethanhe42 over 1 year ago - 9 comments

#9 - HF LLaMA Flax

Issue - State: open - Opened by sanchit-gandhi over 1 year ago - 1 comment

#7 - 13B parameter model

Issue - State: open - Opened by aniquetahir over 1 year ago

#6 - Update

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#5 - Convert back to Hugging Face model

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#4 - Multihost training support

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#3 - Update

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#2 - Update to Llama 2

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago

#1 - Update to Llama 2

Pull Request - State: closed - Opened by ayaka14732 over 1 year ago