INDEX

Explanations

Nothing, as there are no activations above zero to indicate a pattern or preference

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 glim

-0.65

bringer

-0.65

ocracy

-0.65

Pyr

-0.63

 gentleman

-0.63

cohol

-0.62

 convol

-0.62

 anomaly

-0.62

 Humanity

-0.62

 dwar

-0.61

POSITIVE LOGITS

oming

0.77

fu

0.77

omy

0.76

FG

0.75

GE

0.74

âĵĺ

0.72

enne

0.67

 incorrectly

0.66

æĺ

0.63

Â©

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.