INDEX

Explanations

terms indicating popularity or preference for something

oai_token-act-pair · gpt-3.5-turbo

references to favorite things or preferences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

urers

-0.86

ulative

-0.82

ural

-0.77

okin

-0.76

idem

-0.74

ional

-0.74

ene

-0.74

ijk

-0.74

heed

-0.74

OUT

-0.72

POSITIVE LOGITS

 haunt

0.95

 favorites

0.91

 haun

0.86

 favorite

0.84

 underdog

0.78

 darling

0.77

Favorite

0.77

 whipping

0.75

é¾įå¥ĳå£«

0.74

 favourites

0.72

Activations Density 0.032%

terms indicating popularity or preference for something

references to favorite things or preferences

No Comments

No Known Activations