INDEX

Explanations

mentions of importance or significance

oai_token-act-pair · gpt-3.5-turbo

references to the concept of importance

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wonderland

-0.84

TERN

-0.70

 Labyrinth

-0.68

 Flesh

-0.65

 Sheep

-0.64

Snow

-0.63

dream

-0.63

 Flat

-0.62

 Jelly

-0.62

 Monstrous

-0.61

POSITIVE LOGITS

 importance

1.29

 proble

1.13

 significance

0.93

 tremend

0.89

uese

0.87

 traged

0.87

xual

0.86

olicy

0.86

alore

0.85

notation

0.85

Activations Density 0.012%

mentions of importance or significance

references to the concept of importance

No Comments

No Known Activations