INDEX

Explanations

references to different modes and settings in a theater environment

oai_token-act-pair · gpt-3.5-turbo

references to theater or performance modes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

arians

-0.78

ogy

-0.74

loss

-0.69

lag

-0.67

jri

-0.66

 barg

-0.65

 sacrific

-0.64

externalActionCode

-0.64

 retri

-0.63

 mete

-0.62

POSITIVE LOGITS

 Mode

1.05

 Thumbnails

1.00

 Enix

0.85

Exit

0.73

 Created

0.71

ĸļ

0.71

 Virtue

0.69

 Modes

0.68

 Currency

0.66

 Loading

0.65

Activations Density 0.015%

references to different modes and settings in a theater environment

references to theater or performance modes

No Comments

No Known Activations