INDEX

Explanations

references to locations, specifically Washington D.C., and surrounding areas

oai_token-act-pair · gpt-3.5-turbo

instances of references to Washington, D.C. and its significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 mamm

-0.66

ãĤ©

-0.64

 mant

-0.64

 levers

-0.63

 pudding

-0.62

 timet

-0.62

 underpin

-0.62

 proposition

-0.59

 retrospective

-0.59

 chopping

-0.59

POSITIVE LOGITS

Va

1.22

rex

0.94

Gray

0.91

apolis

0.78

0.76

pipe

0.73

trump

0.71

RAW

0.71

 Eisenhower

0.68

wikipedia

0.68

Activations Density 0.039%

references to locations, specifically Washington D.C., and surrounding areas

instances of references to Washington, D.C. and its significance

No Comments

No Known Activations