INDEX

Explanations

mentions of political figures and actions-related to criticism or consequences

oai_token-act-pair · gpt-3.5-turbo

the word "MORE" and its association with political commentary or analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

spring

-0.69

arist

-0.69

ars

-0.66

ãĥ³

-0.65

release

-0.64

bas

-0.63

ãĥª

-0.63

king

-0.63

idem

-0.63

POSITIVE LOGITS

 MORE

1.28

 VIDEOS

0.97

pedia

0.86

osponsors

0.83

FTWARE

0.81

 Flavoring

0.78

eller

0.75

aukee

0.75

 abund

0.74

ellen

0.73

Activations Density 0.006%

mentions of political figures and actions-related to criticism or consequences

the word "MORE" and its association with political commentary or analysis

No Comments

No Known Activations