INDEX

Explanations

references to specific television shows, episodes, events, and health-related topics

oai_token-act-pair · gpt-3.5-turbo

references to popular TV shows and segments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

agara

-0.83

aff

-0.82

uga

-0.81

rum

-0.79

intent

-0.77

umph

-0.75

upp

-0.75

Äĩ

-0.75

internal

-0.75

bon

-0.74

POSITIVE LOGITS

 Initiative

1.11

 Shooter

1.00

 Productions

0.97

 Comics

0.96

 Geek

0.95

 Solutions

0.95

 Tracker

0.94

 Arcade

0.94

 Cafe

0.93

 Baseball

0.93

Activations Density 0.200%

references to specific television shows, episodes, events, and health-related topics

references to popular TV shows and segments

No Comments

No Known Activations