INDEX

Explanations

structure and syntax in coding or programming language

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

user input

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

 resourceCulture

-0.42

rungsseite

-0.41

Resin

-0.39

IntoConstraints

-0.39

PreExecute

-0.37

 çap

-0.36

ⓘ

-0.36

WarningLevel

-0.36

 propOrder

-0.34

DoubleQuotes

-0.34

POSITIVE LOGITS

RegressionTest

0.61

tagHelperRunner

0.53

principalTable

0.52

 AssemblyProduct

0.47

 trial

0.45

mybatisplus

0.42

 thiệu

0.42

Auszeichnungen

0.41

utilisons

0.39

 finale

0.39

Activations Density 0.010%

structure and syntax in coding or programming language

user input

No Comments

No Known Activations