INDEX

Explanations

references to the political figure "Sanders"

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific political figure, Bernie Sanders

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

onyms

-0.78

ocaust

-0.78

omaly

-0.75

obe

-0.74

ozyg

-0.73

othy

-0.72

ocard

-0.71

uclear

-0.71

ophys

-0.71

olulu

-0.71

POSITIVE LOGITS

 Sanders

1.31

Sanders

1.25

 Supporters

0.92

Bernie

0.88

INTON

0.83

 delegates

0.79

 Caucus

0.79

legates

0.79

Malley

0.77

rade

0.76

Activations Density 0.009%

references to the political figure "Sanders"

mentions of a specific political figure, Bernie Sanders

No Comments

No Known Activations