INDEX

Explanations

references to coding and API interactions

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

entering into fields

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

 ligiloj

-0.64

########.

-0.62

 contextLoads

-0.58

 <=",

-0.57

domada

-0.56

EndInit

-0.56

__(/*!

-0.54

 صوتيه

-0.54

 sumpay

-0.54

AxisAlignment

-0.53

POSITIVE LOGITS

WithMany

0.35

Row

0.33

Насе

0.31

doc

0.30

ży

0.30

Runner

0.28

Art

0.28

Context

0.28

Doc

0.28

houding

0.28

Activations Density 0.183%

references to coding and API interactions

entering into fields

No Comments

No Known Activations

references to coding and API interactions

entering into fields

No Comments

No Known Activations