INDEX

Explanations

terms related to software usage and documentation

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

information storage and systems

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

 depending

-0.34

SetOutput

-0.34

nu

-0.34

GUILayout

-0.32

 officially

-0.31

Ein

-0.31

 оригіналу

-0.31

 official

-0.31

 Jost

-0.30

 copy

-0.30

POSITIVE LOGITS

LookAnd

0.91

 informée

0.85

 estekak

0.81

AsUp

0.80

 للمعارف

0.78

RectangleBorder

0.75

 surla

0.74

+#+#

0.71

RTEX

0.71

ValueStyle

0.70

Activations Density 0.006%

terms related to software usage and documentation

information storage and systems

No Comments

No Known Activations