INDEX

Explanations

numerical data and references in the text

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

years

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

oss

-0.46

 arşivlendi

-0.43

клопе

-0.42

 המח

-0.40

med

-0.39

 μά

-0.39

Kole

-0.38

ple

-0.37

 kole

-0.36

 Kole

-0.36

POSITIVE LOGITS

 surla

0.67

reactstrap

0.59

 Савезне

0.58

AccessorTable

0.57

 invokingState

0.51

튿

0.51

principalColumn

0.50

 avoient

0.50

 OFDb

0.50

libatkan

0.48

Activations Density 0.041%

numerical data and references in the text

years

No Comments

No Known Activations