My Argilla
✍
Jean Francois Gutierrez
FranZuzz
·
AI & ML interests
None yet
Recent Activity
published
a Space
7 days ago
FranZuzz/my-argilla
updated
a model
8 months ago
FranZuzz/franzuzz
liked
a dataset
9 months ago
zachgitt/comedy-transcripts
Organizations
None yet
FranZuzz's activity
published
a
Space
7 days ago
updated
a
model
8 months ago
liked
a
dataset
9 months ago
liked
2
models
9 months ago
updated
a
collection
12 months ago
liked
a
model
12 months ago
Interesting! Did you fine tune it with a health/medical dataset ?
free API Interface for Grok
19
#26 opened about 1 year ago
by
ai-leet
120b GGUF quants when?
1
#57 opened about 1 year ago
by
6346y9uey
liked
a
dataset
about 1 year ago
updated
a
Space
about 1 year ago
liked
a
model
about 1 year ago
reacted to
akhaliq's
post with 🤯
about 1 year ago
Post
LongRoPE
Extending LLM Context Window Beyond 2 Million Tokens
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens (2402.13753)
Large context window is a desirable feature in large language models (LLMs). However, due to high fine-tuning costs, scarcity of long texts, and catastrophic values introduced by new token positions, current extended context windows are limited to around 128k tokens. This paper introduces LongRoPE that, for the first time, extends the context window of pre-trained LLMs to an impressive 2048k tokens, with up to only 1k fine-tuning steps at within 256k training lengths, while maintaining performance at the original short context window. This is achieved by three key innovations: (i) we identify and exploit two forms of non-uniformities in positional interpolation through an efficient search, providing a better initialization for fine-tuning and enabling an 8x extension in non-fine-tuning scenarios; (ii) we introduce a progressive extension strategy that first fine-tunes a 256k length LLM and then conducts a second positional interpolation on the fine-tuned extended LLM to achieve a 2048k context window; (iii) we readjust LongRoPE on 8k length to recover the short context window performance. Extensive experiments on LLaMA2 and Mistral across various tasks demonstrate the effectiveness of our method. Models extended via LongRoPE retain the original architecture with minor modifications to the positional embedding, and can reuse most pre-existing optimizations.
Extending LLM Context Window Beyond 2 Million Tokens
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens (2402.13753)
Large context window is a desirable feature in large language models (LLMs). However, due to high fine-tuning costs, scarcity of long texts, and catastrophic values introduced by new token positions, current extended context windows are limited to around 128k tokens. This paper introduces LongRoPE that, for the first time, extends the context window of pre-trained LLMs to an impressive 2048k tokens, with up to only 1k fine-tuning steps at within 256k training lengths, while maintaining performance at the original short context window. This is achieved by three key innovations: (i) we identify and exploit two forms of non-uniformities in positional interpolation through an efficient search, providing a better initialization for fine-tuning and enabling an 8x extension in non-fine-tuning scenarios; (ii) we introduce a progressive extension strategy that first fine-tunes a 256k length LLM and then conducts a second positional interpolation on the fine-tuned extended LLM to achieve a 2048k context window; (iii) we readjust LongRoPE on 8k length to recover the short context window performance. Extensive experiments on LLaMA2 and Mistral across various tasks demonstrate the effectiveness of our method. Models extended via LongRoPE retain the original architecture with minor modifications to the positional embedding, and can reuse most pre-existing optimizations.
- 3 replies
liked
a
Space
about 1 year ago
