INDEX

Explanations

words related to political entities or contexts

oai_token-act-pair · gpt-3.5-turbo

words related to political discussions or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

earances

-0.71

ammy

-0.70

reciation

-0.69

recated

-0.66

 brace

-0.63

 Allaah

-0.63

 Marketable

-0.61

javascript

-0.60

ursed

-0.60

 dunno

-0.59

POSITIVE LOGITS

wagen

1.07

geist

0.75

opal

0.73

ãĤ´ãĥ³

0.72

Cola

0.72

steen

0.72

çĭ

0.71

iform

0.70

Ped

0.69

imaru

0.68

Activations Density 0.161%

words related to political entities or contexts

words related to political discussions or topics

No Comments

No Known Activations