INDEX

Explanations

specific proper nouns and unique identifiers of entities or concepts

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

concrete nouns and proper nouns

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

opsida

-0.41

SequentialGroup

-0.39

TypeCode

-0.38

 становника

-0.37

++.

-0.37

اریخ

-0.35

rbp

-0.35

 inviting

-0.35

 disambiguazione

-0.35

TargetException

-0.34

POSITIVE LOGITS

MigrationBuilder

0.65

RenderAtEndOf

0.52

uxxxx

0.50

SourceChecksum

0.49

ViewImports

0.47

posedge

0.47

 approached

0.46

 enters

0.46

цездатний

0.45

 Мексичка

0.44

Activations Density 0.095%

specific proper nouns and unique identifiers of entities or concepts

concrete nouns and proper nouns

No Comments

No Known Activations