INDEX

Explanations

technical terms related to programming and software structure

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

code structure or identifiers

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

Negative Logits

IntoConstraints

-0.93

 ModelExpression

-0.93

NameInMap

-0.82



-0.80

 propOrder

-0.75

tagHelperRunner

-0.74

RenderAtEndOf

-0.71

MessageOf

-0.67

Personendaten

-0.67

isContained

-0.66

POSITIVE LOGITS

 duradero

0.39

Hallo

0.37

 meyd

0.36

 deseas

0.36

const

0.35

 convivencia

0.35

Hello

0.34

 berguna

0.34

<h1>

0.34

Activations Density 0.088%

technical terms related to programming and software structure

code structure or identifiers

No Known Activations