INDEX

Explanations

phrases indicating knowledge or understanding

oai_token-act-pair · gpt-3.5-turbo

instances of the word "knows."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

phrine

-0.94

oples

-0.77

isco

-0.69

unal

-0.68

OPLE

-0.68

nesota

-0.68

anmar

-0.67

uld

-0.67

interstitial

-0.67

adies

-0.66

POSITIVE LOGITS

ledged

1.04

ledge

0.84

terday

0.78

 whats

0.77

how

0.72

lege

0.71

 exactly

0.71

afer

0.71

how

0.70

LED

0.67

Activations Density 0.043%

phrases indicating knowledge or understanding

instances of the word "knows."

No Comments

No Known Activations