INDEX

Explanations

terms related to biological structures or processes, particularly involving sponges and materials

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

image sensor processing

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

ंदीखरीदारी

-0.78

Personendaten

-0.76

RegressionTest

-0.69

DockStyle

-0.68

 iprot

-0.68

 betweenstory

-0.68

IntoConstraints

-0.67

 surla

-0.67

 تضيفلها

-0.66

 Taktlose

-0.66

POSITIVE LOGITS

 relationship

0.31

sima

0.31

 strong

0.30

 heaviest

0.30

("./

0.30

 anti

0.30

obj

0.30

{}/

0.29

 banking

0.29

 Anti

0.29

Activations Density 0.056%

terms related to biological structures or processes, particularly involving sponges and materials

image sensor processing

No Comments

No Known Activations