INDEX

Explanations

adjectives or nouns denoting deep significance or purpose

oai_token-act-pair · gpt-3.5-turbo

concepts related to the idea of "meaning."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

aliation

-0.69

 Citiz

-0.67

osterone

-0.67

iets

-0.67

asting

-0.66

avorite

-0.65

 Frazier

-0.65

asted

-0.63

DERR

-0.61

emetery

-0.59

POSITIVE LOGITS

fully

1.65

ful

1.38

lessness

1.32

fulness

1.25

lessly

1.02

ual

0.98

sworth

0.96

istically

0.96

FUL

0.95

full

0.92

Activations Density 0.034%

adjectives or nouns denoting deep significance or purpose

concepts related to the idea of "meaning."

No Comments

No Known Activations