INDEX

Explanations

qualitative descriptors indicating positive attributes or evaluations

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

 Cardona

-0.56

 atras

-0.49

Membrane

-0.47

 Montoya

-0.47

 involu

-0.45

 personalisation

-0.45

Enzo

-0.45

 MESSAGE

-0.45

 SONG

-0.45

thalene

-0.44

POSITIVE LOGITS

 fairly

0.85

 pretty

0.82

pretty

0.82

 Fairly

0.73

 assez

0.73

Pretty

0.73

 ganska

0.73

 довольно

0.73

 decent

0.71

 Pretty

0.71

Activations Density 0.184%

qualitative descriptors indicating positive attributes or evaluations

No Comments

No Known Activations