INDEX

Explanations

words related to physical attributes or actions related to rough, uneven surfaces

oai_token-act-pair · gpt-3.5-turbo

references to distribution or dispersion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

externalActionCode

-0.83

displayText

-0.79

yrinth

-0.75

Export

-0.74

ItemTracker

-0.67

Interested

-0.67

Shock

-0.66

States

-0.66

IDS

-0.66

 Loans

-0.66

POSITIVE LOGITS

 stre

1.18

wn

0.83

rol

0.72

Ley

0.68

uously

0.65

cially

0.65

 shredded

0.63

ing

0.62

ricia

0.61

acers

0.61

Activations Density 0.011%

words related to physical attributes or actions related to rough, uneven surfaces

references to distribution or dispersion

No Comments

No Known Activations