INDEX

Explanations

references to programming components and frameworks

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

technical object names

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

 strict

-0.31

 relaxed

-0.29

หว

-0.29

↵↵

-0.28

-0.27

来

-0.26

 Stamm

-0.26

 medium

-0.26

grand

-0.25

 برد

-0.25

POSITIVE LOGITS

RenderAtEndOf

0.95

 ProtoMessage

0.79

MigrationBuilder

0.77

Diweddarwch

0.74

 müſſen

0.69

 AssemblyCulture

0.68

 typelib

0.68

ſelben

0.67

 kasarigan

0.66

ValueStyle

0.66

Activations Density 0.249%

references to programming components and frameworks

technical object names

No Comments

No Known Activations