INDEX

Explanations

key phrases indicating future events or planning

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

auxiliary verbs followed by verbs

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

ngdoc

-0.57

RegressionTest

-0.52

DoubleQuotes

-0.52

InstrumentedTest

-0.49

VersionUID

-0.48

GeoNames

-0.46

MarshalTo

-0.45

WriteAttribute

-0.45

elemField

-0.41

دانشنامهٔ

-0.41

POSITIVE LOGITS

êmica

0.41

ラク

0.41

AddTagHelper

0.41

లాలు

0.40

lichkeit

0.40

deed

0.39

liğini

0.38

liği

0.37

 utiles

0.37

leit

0.37

Activations Density 0.543%

key phrases indicating future events or planning

auxiliary verbs followed by verbs

No Comments

No Known Activations