INDEX

Explanations

adjectives describing negative traits or behaviors

oai_token-act-pair · gpt-3.5-turbo

abstract nouns that describe negative qualities or conditions

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

oval

-0.76

ãĤ¤ãĥĪ

-0.66

iverse

-0.65

ãĥİ

-0.65

 Colon

-0.65

ãĥ´

-0.64

oret

-0.63

inav

-0.63

ramer

-0.62

 Federation

-0.62

POSITIVE LOGITS

fulness

1.12

iness

1.00

ously

0.96

lessness

0.88

lust

0.88

 ridden

0.85

chery

0.83

uous

0.83

ulence

0.82

FUL

0.82

Activations Density 0.137%

adjectives describing negative traits or behaviors

abstract nouns that describe negative qualities or conditions

No Comments

No Known Activations