INDEX

Explanations

code-related structures and syntax in programming or markup languages

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

language variety

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

 estekak

-0.75

 ujednoznacz

-0.68

 CreateTagHelper

-0.63

 informée

-0.61

__(/*!

-0.61

 المعيارى

-0.61

ContentAsync

-0.61

出版年

-0.60

AutoScaleMode

-0.60

原始内容存档于

-0.59

POSITIVE LOGITS

ENCODING

0.36

 Folge

0.35

 limitado

0.35

 caído

0.35

 encuentre

0.34

 līdz

0.34

 haría

0.33

ADORA

0.33

La

0.33

 darbu

0.33

Activations Density 0.118%

code-related structures and syntax in programming or markup languages

language variety

No Comments

No Known Activations