INDEX

Explanations

references to personally identifiable information and data privacy practices

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

personal information

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

帖最后由

-0.61

 ivelany

-0.59

ніципалі

-0.53

IUrlHelper

-0.52

expandindo

-0.51

فایل‌لار

-0.50

лтамалар

-0.50

Viited

-0.50

 Савезне

-0.49

 defaultstate

-0.49

POSITIVE LOGITS

 information

0.42

了一眼

0.36

 MONEY

0.36

 personali

0.36

ModelSerializer

0.36

 mapStateToProps

0.35

internalType

0.35

情報は

0.35

Únete

0.35

 SIGN

0.35

Activations Density 0.046%

references to personally identifiable information and data privacy practices

personal information

No Comments

No Known Activations