INDEX

Explanations

people's names or specific terms related to individuals or entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns and references to specific entities or individuals

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

 Dickinson

-0.67

 cents

-0.64

hed

-0.61

 descriptor

-0.58

 gratification

-0.56

rpm

-0.56

 signalling

-0.55

 payoff

-0.55

 Robbins

-0.55

 horsepower

-0.52

POSITIVE LOGITS

Äĩ

0.86

sbm

0.79

Downloadha

0.75

achus

0.72

anmar

0.67

 Parish

0.65

 Torment

0.65

angan

0.63

gmail

0.63

Ã¡n

0.63

Activations Density 1.053%

people's names or specific terms related to individuals or entities

proper nouns and references to specific entities or individuals

No Comments

No Known Activations