INDEX

Explanations

phrases related to legal or licensing terms

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

controlled through or around

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

例句

-0.50

➌

-0.48

 varones

-0.43

Litteratur

-0.42

SourceChecksum

-0.42

 despedida

-0.41

 Moscú

-0.40

ciorys

-0.40

cellation

-0.39

 třeba

-0.39

POSITIVE LOGITS

 nahilalakip

0.68

 MainAxisSize

0.59

 мәкал

0.55

AddTagHelper

0.55

 control

0.54

 controlling

0.52

AccessorTable

0.52

ConstraintMaker

0.51

 Controlled

0.50

 Control

0.49

Activations Density 0.000%

phrases related to legal or licensing terms

controlled through or around

No Comments

No Known Activations