INDEX

Explanations

keywords related to specific incidents or events

oai_token-act-pair · gpt-3.5-turbo

references to events and incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

intel

-0.64

anol

-0.64

ixel

-0.63

regon

-0.62

 "$:/

-0.59

dust

-0.57

irtual

-0.57

ingers

-0.56

erd

-0.56

 Shooter

-0.55

POSITIVE LOGITS

 circulated

0.74

 surfaced

0.70

 concluded

0.69

 aired

0.68

>>>>>>>>

0.68

cens

0.67

 attest

0.62

 unfold

0.61

 CLICK

0.60

 concludes

0.60

Activations Density 0.525%

keywords related to specific incidents or events

references to events and incidents

No Comments

No Known Activations