INDEX

Explanations

texts related to documents or written information

oai_token-act-pair · gpt-3.5-turbo

references to transcripts and detailed descriptions of content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

aneers

-1.03

alties

-0.98

storms

-0.88

entimes

-0.87

ampires

-0.87

ornings

-0.82

auts

-0.79

ãĥĺ

-0.78

asures

-0.78

asions

-0.77

POSITIVE LOGITS

 rundown

1.20

 breakdown

1.17

 kicker

1.16

 synopsis

1.14

 excerpt

1.06

 link

1.04

 gist

1.01

 backstory

1.00

 thing

0.96

 list

0.96

Activations Density 0.107%

texts related to documents or written information

references to transcripts and detailed descriptions of content

No Comments

No Known Activations