INDEX

Explanations

phrases related to political figures or discussions

oai_token-act-pair · gpt-3.5-turbo

specific characters or symbols and their usage in a textual context

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

 seiz

-0.74

 horizont

-0.70

bda

-0.64

resses

-0.64

 comprom

-0.63

iae

-0.63

 flex

-0.63

othal

-0.63

IPM

-0.62

 clicks

-0.62

POSITIVE LOGITS

gon

0.86

Connor

0.85

culus

0.84

Malley

0.81

Wan

0.80

Donnell

0.79

minecraft

0.78

gress

0.78

ternity

0.76

Neill

0.76

Activations Density 0.055%

phrases related to political figures or discussions

specific characters or symbols and their usage in a textual context

No Comments

No Known Activations