INDEX

Explanations

code-related structure openings and specific formatting markers

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

 onData

-0.36

 docente

-0.32

 setText

-0.31

zieher

-0.30

 trò

-0.30

虎

-0.29

 geografia

-0.29

Resource

-0.29

citizen

-0.29

 saites

-0.29

POSITIVE LOGITS

GEBURTSDATUM

0.88

RegressionTest

0.76

uxxxx

0.73

帖最后由

0.69

fromnode

0.68

 tartalomajánló

0.66

 يتيمه

0.66

transQ

0.65

richTextPanel

0.65

 GenerationType

0.64

Activations Density 1.625%

code-related structure openings and specific formatting markers

No Comments

No Known Activations