INDEX

Explanations

identifiers and parameters used in programming or data configuration contexts

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

code identifiers and values

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

 CreateTagHelper

-0.68

ViewFeatures

-0.63

GraphicsUnit

-0.59

-0.56

ChrTalk

-0.56

SPATH

-0.55

richTextPanel

-0.54

ViewImports

-0.50

 estekak

-0.50

 informée

-0.50

POSITIVE LOGITS

HandlerContext

0.42

chartInstance

0.40

HttpDelete

0.39

RegressionTest

0.38

ścio

0.36

出版年

0.35

IsMutable

0.35

ArrowToggle

0.34

 sqlSession

0.34

aronder

0.33

Activations Density 0.140%

identifiers and parameters used in programming or data configuration contexts

code identifiers and values

No Comments

No Known Activations