INDEX

Explanations

programming functions related to data manipulation and thread management

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

program code snippets

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

ปก

-0.34

 conclusiones

-0.32

featureID

-0.32

จบ

-0.32

IDLE

-0.31

 klaar

-0.30

 Ensino

-0.30

illac

-0.30

 fabrics

-0.29

 protección

-0.29

POSITIVE LOGITS

astify

0.70

Personendaten

0.65

ItemBackground

0.62

 للمعارف

0.56

richTextPanel

0.55

 دیکھیے

0.52

 Autorisations

0.50

WriteTagHelper

0.49

inSlope

0.49

ternut

0.49

Activations Density 0.029%

programming functions related to data manipulation and thread management

program code snippets

No Comments

No Known Activations