INDEX

Explanations

references to user-related actions and behaviors in coding contexts

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

conditions or comparisons

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 apre

-0.39

ESE

-0.39

setupUi

-0.38

IntoConstraints

-0.38

 Olms

-0.38

ese

-0.38

rek

-0.36

喜

-0.36

ServerSocket

-0.35

HT

-0.35

POSITIVE LOGITS

 ویکی‌پدی

0.72

 informée

0.57

Jeografia

0.54

 desmotivaciones

0.54

 kasarigan

0.51

şört

0.51

lgari

0.50

rungsseite

0.50

 kontakta

0.49

setNome

0.48

Activations Density 0.001%

references to user-related actions and behaviors in coding contexts

conditions or comparisons

No Comments

No Known Activations