INDEX

Explanations

questions starting with "How can" or "How could"

oai_token-act-pair · gpt-3.5-turbo

questions that express confusion or doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 vert

-0.70

Board

-0.70

 simulator

-0.63

 skating

-0.62

board

-0.62

ASC

-0.58

 Dickinson

-0.57

boarding

-0.55

beard

-0.55

 ported

-0.55

POSITIVE LOGITS

berra

1.07

't

1.05

ataka

0.80

adian

0.74

NOT

0.74

ADA

0.72

ameless

0.70

Correct

0.70

iably

0.70

]);

0.69

Activations Density 0.032%

questions starting with "How can" or "How could"

questions that express confusion or doubt

No Comments

No Known Activations