INDEX

Explanations

proper nouns related to locations or entities

oai_token-act-pair · gpt-3.5-turbo

references to the location "Waterloo" and related names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

idency

-0.77

hedon

-0.70

inging

-0.69

aic

-0.68

ing

-0.68

ospace

-0.66

amental

-0.65

eport

-0.65

EMENT

-0.65

 Strauss

-0.65

POSITIVE LOGITS

 Klux

0.88

NP

0.71

Jet

0.69

jet

0.69

 retri

0.68

Ļ½

0.67

Ont

0.66

 Typhoon

0.65

naires

0.64

Gaw

0.64

Activations Density 0.048%

proper nouns related to locations or entities

references to the location "Waterloo" and related names

No Comments

No Known Activations