INDEX

Explanations

phrases indicating the presentation or outlining of information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating plans, proposals, or arguments being presented

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 artifact

-0.65

ni

-0.64

 luck

-0.63

 unnoticed

-0.61

inos

-0.60

 nose

-0.60

Ã£o

-0.59

blers

-0.59

avery

-0.58

iol

-0.57

POSITIVE LOGITS

 outlines

1.13

\\\\\\\\

0.95

 outline

0.91

 outlining

0.77

 Goals

0.75

how

0.75

WER

0.74

 guidelines

0.73

 plans

0.72

plan

0.70

Activations Density 0.077%

phrases indicating the presentation or outlining of information

phrases indicating plans, proposals, or arguments being presented

No Comments

No Known Activations