INDEX

Explanations

chemical or programming-related terminology

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

pellets and coefficients

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 ElementRef

-0.42

 Deli

-0.41

specialchars

-0.41

 *)__

-0.41

DoubleQuotes

-0.40

 AppCompat

-0.40

Já

-0.39

Victim

-0.39

 Burn

-0.39

itemid

-0.38

POSITIVE LOGITS

inSlope

0.89

piram

0.54

addCriterion

0.50

httphttps

0.47

NamedQuery

0.47

SharedCtor

0.46

0.45

yych

0.45

makeText

0.44

SequentialGroup

0.42

Activations Density 0.013%

chemical or programming-related terminology

pellets and coefficients

No Comments

No Known Activations