INDEX

Explanations

code-related comments and documentation within programming language syntax

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

HTML emphasis tags and http

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

 queſta

-0.98

featureID

-0.98

parsedMessage

-0.92

 Biôgrafia

-0.89

ðsíða

-0.85

ロウィン

-0.84

 ویکی‌پدی

-0.81

majánló

-0.78

ſammen

-0.78

 Мексичка

-0.77

POSITIVE LOGITS

↵

0.53

0.44

<em>

0.42

0.41

IN

0.41

<i>

0.41

section

0.41

http

0.41

Activations Density 0.003%

code-related comments and documentation within programming language syntax

HTML emphasis tags and http

No Comments

No Known Activations