INDEX

Explanations

phrases indicating independence or conditions affecting outcomes

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

regardless of condition

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

RenderAtEndOf

-0.54

 GenerationType

-0.49

expliquer

-0.47

 لينك

-0.46

 blancas

-0.46

 CURIAM

-0.45

BeginContext

-0.44

ChrTalk

-0.44

 kollu

-0.44

%)$

-0.44

POSITIVE LOGITS

propTypes

0.44

onViewCreated

0.40

WebControls

0.39

defaultProps

0.36

IBAction

0.35

createStatement

0.35

routeProvider

0.35

setupUi

0.34

Ligações

0.33

Eksterne

0.33

Activations Density 0.092%

phrases indicating independence or conditions affecting outcomes

regardless of condition

No Comments

No Known Activations

phrases indicating independence or conditions affecting outcomes

regardless of condition

No Comments

No Known Activations