INDEX

Explanations

images with captions that have been enlarged

oai_token-act-pair · gpt-3.5-turbo

instances of the word "this."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

bats

-0.75

-0.67

aturdays

-0.61

 fronts

-0.61

termination

-0.60

spring

-0.57

 affairs

-0.57

 Tend

-0.57

naires

-0.56

 lobb

-0.56

POSITIVE LOGITS

 image

1.00

 ARTICLE

0.86

 Image

0.81

 toggle

0.77

image

0.76

ption

0.72

embed

0.72

malink

0.72

 slide

0.70

 Advertisement

0.70

Activations Density 0.008%

images with captions that have been enlarged

instances of the word "this."

No Comments

No Known Activations