INDEX

Explanations

error corrections in news articles

oai_token-act-pair · gpt-3.5-turbo

corrections or clarifications to previously stated information

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Mods

-0.76

Ń·

-0.69

mods

-0.66

SpaceEngineers

-0.64

abe

-0.63

¥µ

-0.62

 Cthulhu

-0.61

Frameworks

-0.61

Plex

-0.61

soDeliveryDate

-0.61

POSITIVE LOGITS

 spelling

0.86

 incorrectly

0.85

 corrected

0.85

 typo

0.84

 incorrect

0.82

 mistakenly

0.80

Clar

0.78

 erroneous

0.75

 cler

0.74

 headline

0.74

Activations Density 0.143%

error corrections in news articles

corrections or clarifications to previously stated information

No Comments

No Known Activations