INDEX

Explanations

phrases emphasizing the large or total quantity of something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating universality or generalization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ubi

-0.76

uel

-0.67

only

-0.63

yna

-0.63

uers

-0.61

alin

-0.61

kind

-0.60

 Balls

-0.60

most

-0.60

 Dull

-0.59

POSITIVE LOGITS

 imaginable

0.88

WHERE

0.87

THING

0.76

hyde

0.75

 conceivable

0.75

ä»£

0.65

Ùĩ

0.64

body

0.63

 extant

0.62

 else

0.62

Activations Density 0.112%

phrases emphasizing the large or total quantity of something

phrases indicating universality or generalization

No Comments

No Known Activations