INDEX

Explanations

ethnicity-related adjectives paired with nationality terms

oai_token-act-pair · gpt-3.5-turbo

terms related to various social and political identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ravings

-1.12

ologies

-1.10

ories

-1.04

encies

-1.00

boards

-0.98

renches

-0.97

poons

-0.96

gments

-0.95

abilities

-0.95

isites

-0.94

POSITIVE LOGITS

who

1.13

 thinker

1.13

 politician

1.11

 businessman

1.09

 woman

1.03

 traveler

1.02

 adventurer

1.02

 entrepreneur

1.01

 monk

1.01

 athlete

1.01

Activations Density 0.289%

ethnicity-related adjectives paired with nationality terms

terms related to various social and political identities

No Comments

No Known Activations