INDEX

Explanations

the word 'series'

oai_token-act-pair · gpt-4-turbo

mentions of TV or book series

oai_token-act-pair · gpt-3.5-turbo

references to television or animated series

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lig

-0.79

lite

-0.78

nai

-0.76

ryn

-0.73

achev

-0.72

lesi

-0.72

imentary

-0.71

orget

-0.71

anged

-0.70

ossier

-0.70

POSITIVE LOGITS

ãĥĺ

0.96

ĸļ

0.80

manship

0.79

 finale

0.78

enegger

0.75

 è£ıè¦ļéĨĴ

0.72

isodes

0.70

ends

0.69

 Presents

0.69

series

0.68

Activations Density 0.024%

the word 'series'

mentions of TV or book series

references to television or animated series

No Comments

No Known Activations