INDEX

Explanations

mentions of specific publications, especially "New York Daily News"

oai_token-act-pair · gpt-3.5-turbo

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

Äĩ

-0.85

peat

-0.78

blem

-0.76

kaya

-0.71

BIL

-0.71

henko

-0.69

BILITIES

-0.66

vic

-0.65

olate

-0.65

TTL

-0.65

POSITIVE LOGITS

anders

0.65

ptroller

0.63

ository

0.61

Library

0.61

Arcade

0.60

 Booth

0.60

enhagen

0.60

Thumbnail

0.59

chel

0.59

apest

0.59

Activations Density 0.121%

mentions of specific publications, especially "New York Daily News"

No Comments

No Known Activations