INDEX

Explanations

phrases related to extraction or removal

oai_token-act-pair · gpt-3.5-turbo

references to specific entities or key terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pg

-0.79

meal

-0.78

ugh

-0.77

ugi

-0.75

hire

-0.72

uge

-0.70

efully

-0.70

arge

-0.69

FY

-0.69

opic

-0.68

POSITIVE LOGITS

BuyableInstoreAndOnline

0.76

 unlaw

0.72

 Ambro

0.68

(_

0.67

Steam

0.64

RANT

0.64

jriwal

0.64

 loopholes

0.64

 contradictions

0.62

andro

0.61

Activations Density 0.000%

phrases related to extraction or removal

references to specific entities or key terms

No Comments

No Known Activations