INDEX

Explanations

structural components or sections in academic or technical documents

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

"{#...}"

np_acts-logits-general · gemini-2.0-flash

names starting with Rod

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 utafitiHapana

-0.78

Хьажоргаш

-0.66

migrationBuilder

-0.65

/\.

-0.64

-0.62

oredCriteria

-0.60

/*

-0.60

 '\\;'

-0.59

GIVEREF

-0.59

ніципа

-0.58

POSITIVE LOGITS

[::-

0.55

consulté

0.50

itizing

0.46

 Turk

0.45

 Arjun

0.45

discussed

0.44

ulaski

0.43

 RSSSF

0.42

Viungo

0.42

 jeste

0.42

Activations Density 0.073%

structural components or sections in academic or technical documents

"{#...}"

names starting with Rod

No Comments

No Known Activations