INDEX

Explanations

phrases indicating degree or extent

oai_token-act-pair · gpt-3.5-turbo

phrases that discuss varying degrees of impact or influence

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

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

tty

-0.67

brook

-0.66

 Scholar

-0.65

reth

-0.64

 Shal

-0.64

@#&

-0.63

rium

-0.63

SPR

-0.62

«

-0.62

RY

-0.62

POSITIVE LOGITS

 restitution

0.70

mma

0.70

 approximation

0.69

uracy

0.68

iece

0.68

olit

0.66

esm

0.62

 extent

0.62

tenance

0.61

iter

0.61

Activations Density 0.023%

phrases indicating degree or extent

phrases that discuss varying degrees of impact or influence

No Comments

No Known Activations