INDEX

Explanations

references to specific events or actions occurring over time

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

 juſ

-0.59

 themſelves

-0.54

 Monfieur

-0.53

 Conſ

-0.52

 незавершена

-0.50

 quæ

-0.50

 itſelf

-0.50

AndEndTag

-0.50

 himſelf

-0.49

 myſelf

-0.49

POSITIVE LOGITS

 again

1.04

Again

0.88

again

0.86

 Again

0.84

 AGAIN

0.76

 опять

0.75

 novamente

0.71

 opět

0.71

 nuevamente

0.69

 igjen

0.68

Activations Density 0.485%

references to specific events or actions occurring over time

No Comments

No Known Activations