INDEX

Explanations

references to the State Department

oai_token-act-pair · gpt-3.5-turbo

references to the State Department

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ°

-0.76

rights

-0.73

 sidx

-0.73

stre

-0.70

 Pose

-0.69

iser

-0.69

tackle

-0.68

moving

-0.66

amaz

-0.65

ties

-0.65

POSITIVE LOGITS

artment

0.87

 secretary

0.82

 Department

0.81

 secretaries

0.80

artments

0.80

 Secretary

0.79

 Dept

0.78

 heav

0.74

anooga

0.72

 Intelligence

0.72

Activations Density 0.021%

references to the State Department

references to the State Department

No Comments

No Known Activations