INDEX

Explanations

instances of people being observed or noticed doing various activities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seen."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

idity

-0.72

angan

-0.69

iets

-0.68

 Clicker

-0.63

RET

-0.63

ulum

-0.62

 Nightmares

-0.61

 Correction

-0.60

Newsletter

-0.60

 belief

-0.58

POSITIVE LOGITS

 hovering

0.84

dust

0.79

 photograp

0.78

agascar

0.77

 roaming

0.74

ivating

0.73

ById

0.72

inburgh

0.72

 flickering

0.71

 briefly

0.70

Activations Density 0.052%

instances of people being observed or noticed doing various activities

instances of the word "seen."

No Comments

No Known Activations