INDEX

Explanations

mentions of animated media

oai_token-act-pair · gpt-3.5-turbo

references to animated films or shows

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

orr

-0.84

yer

-0.83

mith

-0.82

hood

-0.76

ILA

-0.75

OA

-0.73

ills

-0.72

Availability

-0.70

lay

-0.70

anship

-0.70

POSITIVE LOGITS

GIF

0.97

ebted

0.81

gif

0.80

 animated

0.79

ocument

0.78

atable

0.75

netflix

0.73

 Animated

0.69

 versions

0.69

 cartoon

0.68

Activations Density 0.017%

mentions of animated media

references to animated films or shows

No Comments

No Known Activations