INDEX

Explanations

technical or programming-related terms and concepts

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

Java and JavaScript code snippets

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

testify

-0.47

TextWatcher

-0.43

 PRESENTED

-0.43

 détaillée

-0.43

artifactId

-0.41

ContentAlignment

-0.40

elemField

-0.39

comentario

-0.39

 adelantado

-0.39

 encre

-0.39

POSITIVE LOGITS

 utafitiHapana

0.46

ScopeManager

0.41

 Roskov

0.39

;->

0.38

autaire

0.37

الدراسه

0.36

StoryboardSegue

0.36

 GlobalKey

0.36

 pylint

0.36

애

0.35

Activations Density 0.051%

technical or programming-related terms and concepts

Java and JavaScript code snippets

No Comments

No Known Activations