INDEX

Explanations

mentions of the animal "lion" within text

oai_token-act-pair · gpt-3.5-turbo

references to lions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

mble

-0.83

lying

-0.79

ilk

-0.79

ACTION

-0.70

Ñı

-0.69

chell

-0.69

 McKenna

-0.68

kson

-0.67

 Morse

-0.65

ADRA

-0.64

POSITIVE LOGITS

esses

1.27

 lions

1.17

fish

1.10

 lion

1.05

ess

1.01

eye

1.00

osaurs

0.94

ous

0.92

odon

0.89

iasis

0.87

Activations Density 0.010%

mentions of the animal "lion" within text

references to lions

No Comments

No Known Activations