INDEX

Explanations

GIF files or mentions of GIFs

oai_token-act-pair · gpt-3.5-turbo

references to GIFs or animated images

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arios

-0.80

ammy

-0.65

Pax

-0.61

heid

-0.61

ologically

-0.61

 Austral

-0.59

aeda

-0.59

anwhile

-0.59

ally

-0.58

chedel

-0.57

POSITIVE LOGITS

TED

1.11

GIF

0.97

TING

0.91

TION

0.91

gif

0.86

TEXT

0.84

RAY

0.82

eries

0.81

 Thumbnails

0.78

FIX

0.77

Activations Density 0.012%

GIF files or mentions of GIFs

references to GIFs or animated images

No Comments

No Known Activations