INDEX

Explanations

mentions of functionalities and features in software or applications

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

features and functionality

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

 chiare

-0.38

 PyTuple

-0.35

openConnection

-0.35

 avaliações

-0.35

 painted

-0.35

mael

-0.34

 investing

-0.34

issa

-0.33

 Estadual

-0.33

 East

-0.33

POSITIVE LOGITS

 functionalities

0.90

 Functionality

0.88

 functionality

0.83

功能

0.81

 Capabilities

0.80

的功能

0.77

 Functions

0.76

 기능

0.75

 հղումներ

0.75

 capabilities

0.74

Activations Density 0.037%

mentions of functionalities and features in software or applications

features and functionality

No Comments

No Known Activations