INDEX

Explanations

mathematical expressions and operations related to lengths and dimensions

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

code blocks and lengths

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

 defaultstate

-0.50

setcounter

-0.49

("")]

-0.44



-0.43

">//

-0.42

++)

-0.42

versa

-0.41

Portail

-0.41

assertIn

-0.41

()));

-0.40

POSITIVE LOGITS

 psicológico

0.51

 fédé

0.48

 cáps

0.47

 térmico

0.47

 '\\;'

0.45

 sportifs

0.44

 prochaines

0.44

tangentMode

0.44

 հղումներ

0.43

 diferencial

0.43

Activations Density 0.021%

mathematical expressions and operations related to lengths and dimensions

code blocks and lengths

No Comments

No Known Activations