INDEX

Explanations

references to existence and living experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-0.59

 Bulldogs

-0.53

 Falcons

-0.52

ünster

-0.52

 others

-0.52

Chandler

-0.49

 chitar

-0.49

 Shorts

-0.49

arities

-0.49

 Chandler

-0.49

POSITIVE LOGITS

 living

0.89

Living

0.85

living

0.85

 Living

0.81

 LIVING

0.76

 Lived

0.75

LIVING

0.73

 lived

0.71

 Live

0.69

Live

0.69

Activations Density 0.091%

references to existence and living experiences

No Comments

No Known Activations