INDEX

Explanations

references to animation, animators, and animated content

oai_token-act-pair · gpt-3.5-turbo

references to animation and animated content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

holder

-0.95

ouston

-0.68

ledge

-0.67

 Norton

-0.66

inness

-0.66

holders

-0.65

ership

-0.65

ructure

-0.64

umbing

-0.63

ront

-0.63

POSITIVE LOGITS

GIF

1.14

gif

0.98

oji

0.82

 studios

0.82

Preview

0.80

agus

0.79

 studio

0.79

 animation

0.76

 shorts

0.75

 film

0.74

Activations Density 0.055%

references to animation, animators, and animated content

references to animation and animated content

No Comments

No Known Activations