INDEX

Explanations

references to closures and obstacles

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

phrases related to closing or ending something

oai_token-act-pair · gpt-3.5-turbo

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

Magikarp

-0.66

 Terr

-0.61

PET

-0.60

Gry

-0.59

Kings

-0.59

Ã±

-0.58

 Lawyers

-0.58

kh

-0.57

 guesses

-0.57

FactoryReloaded

-0.57

POSITIVE LOGITS

 doors

0.96

 arteries

0.86

 door

0.84

 airspace

0.78

 curtains

0.77

rency

0.75

 seams

0.72

 artery

0.72

 Doors

0.72

 prematurely

0.71

Activations Density 0.088%

references to closures and obstacles

phrases related to closing or ending something

No Comments

No Known Activations