johnsmith0031/alpaca_lora_4bit
View on GitHub ↗ ★ 533 Python
8 mentions on Hacker News by 2 people: 1 story, 7 comments · first 2023-03-23 · latest 2023-05-04
20082026
-
About 15GB training it in the webui. If you use https://github.com/johnsmith0031/alpaca_lora_4bit then 30B only needs 24GB, and works on a single 3090 or $200 P40.
comment by MacsHeadroom · 2023-05-04 · in Google “We have no moat, and neither does OpenAI” -
…of hours on any consumer GPU with 8GB of VRAM and release it under the commercial license (using https://github.com/johnsmith0031/alpaca_lora_4bit ).
comment by MacsHeadroom · 2023-04-20 · in StableLM: A new open-source language model -
…any consumer GPU with 8GB of VRAM in a couple of hours and it will be commercial licensed. (using https://github.com/johnsmith0031/alpaca_lora_4bit ) You can even use the exact same dataset StabilityAI used. (Although there are better…
comment by MacsHeadroom · 2023-04-20 · in StableLM: A new open-source language model -
…licensed and you can finetune it in a couple of hours on a consumer GPU with 8GB of VRAM using https://github.com/johnsmith0031/alpaca_lora_4bit
comment by MacsHeadroom · 2023-04-20 · in StableLM: A new open-source language model -
You can finetune 7B in a couple of hours on a $200 3060 with https://github.com/johnsmith0031/alpaca_lora_4bit
comment by MacsHeadroom · 2023-04-20 · in StableLM: A new open-source language model -
…time by orders of magnitude. Here's an example of 20 seconds per epoch on a single consumer GPU: https://github.com/johnsmith0031/alpaca_lora_4bit/issues/7#i...
comment by MacsHeadroom · 2023-03-26 · in LoRA: Low-Rank Adaptation of Large Language Models -
As of yesterday you can train 33B parameter models on a single consumer 24GB VRAM GPU with https://github.com/johnsmith0031/alpaca_lora_4bit . It's already proven that 13B parameters is enough to beat GPT-3 175B quality. It's likely that…
comment by MacsHeadroom · 2023-03-24 · in RWKV RNN: Better than ChatGPT?