INDEX

Explanations

phrases starting with "For those of you" and providing information or explanations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the provision of information or guidance to the reader

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

izons

-0.79

ecided

-0.77

ileged

-0.74

cible

-0.69

ifferent

-0.68

iar

-0.68

ifference

-0.67

etermined

-0.67

rouse

-0.66

vati

-0.65

POSITIVE LOGITS

BAT

0.86

famous

0.66

BALL

0.65

 ãĥ

0.64

éĩ

0.62

BAT

0.62

 Basically

0.61

 acronym

0.61

 Anon

0.60

fuck

0.60

Activations Density 0.504%

phrases starting with "For those of you" and providing information or explanations

phrases indicating the provision of information or guidance to the reader

No Comments

No Known Activations