INDEX

Explanations

explicit expressions of opinions or attitudes

oai_token-act-pair · gpt-3.5-turbo

phrases that express disregard or apathy

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

iba

-0.72

Ct

-0.70

ADE

-0.65

yrinth

-0.64

Link

-0.63

indle

-0.63

_-

-0.62

Downloadha

-0.60

amiya

-0.59

CHA

-0.59

POSITIVE LOGITS

 impression

0.94

 advice

0.75

 condolences

0.75

 impetus

0.74

 insight

0.73

 indication

0.70

icum

0.70

nod

0.69

 preferential

0.69

 keynote

0.69

Activations Density 0.157%

explicit expressions of opinions or attitudes

phrases that express disregard or apathy

No Comments

No Known Activations