INDEX

Explanations

elements indicating additional information or references to external content

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

links and continuation indicators

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

InstrumentedTest

-0.52

EndContext

-0.51

tagHelperRunner

-0.47

AnimationsModule

-0.46

CppMethod

-0.45

الحياه

-0.44

DoubleQuotes

-0.42

κτηρισ

-0.42

TargetException

-0.42

lename

-0.42

POSITIVE LOGITS

verifyException

0.53

цездатний

0.46

 nahilalakip

0.45

 resourceCulture

0.44

$_['

0.42

 EconPapers

0.42

FunctionFlags

0.42

 autorytatywna

0.40

 صوتيه

0.39

 AssemblyCompany

0.37

Activations Density 0.009%

elements indicating additional information or references to external content

links and continuation indicators

No Comments

No Known Activations