INDEX

Explanations

the word "Zen" in the text.

oai_token-act-pair · gpt-3.5-turbo

references to "Zen" as a theme or concept

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

WAYS

-0.79

âĢ¢âĢ¢âĢ¢âĢ¢

-0.75

à¤

-0.74

å¥³

-0.74

riers

-0.71

OTT

-0.70

,,,,,,,,

-0.69

à¨

-0.68

Ø©

-0.67

 Dise

-0.67

POSITIVE LOGITS

geist

0.91

obia

0.87

ith

0.86

eca

0.86

onite

0.86

zen

0.84

ajo

0.84

emy

0.83

jet

0.81

ices

0.81

Activations Density 0.007%

the word "Zen" in the text.

references to "Zen" as a theme or concept

No Comments

No Known Activations