INDEX

Explanations

punctuation marks and formatting indicators in textual data

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

specific technical terms or phrases

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

 Normdatei

-0.85

 propOrder

-0.72

Autoritní

-0.68

MethodManager

-0.67

RTEX

-0.65

новништво

-0.62

UrlResolution

-0.62

⟬

-0.60

GOTREF

-0.58

 OFDb

-0.58

POSITIVE LOGITS

BufferException

0.32

otimes

0.31

Bes

0.30

atiche

0.28

artist

0.28

 QUOTE

0.28

Ant

0.27

ﬅ

0.27

வில்

0.27

 Dist

0.27

Activations Density 0.137%

punctuation marks and formatting indicators in textual data

specific technical terms or phrases

No Comments

No Known Activations