INDEX

Explanations

formatted string outputs or print statements in code

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

multilingual technical code

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

Numerade

-0.34

 ✭✭

-0.33

 aans

-0.33

fire

-0.32

 diser

-0.32

links

-0.32

 дали

-0.32

ósz

-0.32

stdc

-0.31

NSIndexPath

-0.31

POSITIVE LOGITS

تقاوى

0.69

 виправивши

0.67

AddHtmlAttribute

0.62

最快更新

0.57

 resourceCulture

0.55

 تضيفلها

0.54

下载附件

0.52

 JpaRepository

0.50

новниш

0.50

 estekak

0.49

Activations Density 0.007%

formatted string outputs or print statements in code

multilingual technical code

No Comments

No Known Activations