INDEX

Explanations

references to the color yellow

oai_token-act-pair · gpt-3.5-turbo

references to the color yellow

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

SPONSORED

-0.85

inion

-0.78

RT

-0.72

itia

-0.71

VICE

-0.69

REL

-0.69

 TODAY

-0.68

Prosecut

-0.67

Engine

-0.66

Far

-0.65

POSITIVE LOGITS

 yellow

1.07

 coloured

0.99

yellow

0.95

cake

0.91

 colored

0.87

 oval

0.87

 tint

0.87

 pigment

0.86

 pink

0.84

 Yellow

0.83

Activations Density 0.007%

references to the color yellow

references to the color yellow

No Comments

No Known Activations