INDEX

Explanations

terms related to background checks and criminal history records

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

background checks and clearance

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

findpost

-0.75

SourceChecksum

-0.68

classnames

-0.68

帖最后由

-0.59

usermodel

-0.58

httphttps

-0.58

SequentialGroup

-0.58

OrNil

-0.57

expandindo

-0.56



-0.56

POSITIVE LOGITS

 enquête

0.35

 sangre

0.34

Hentet

0.34

 contratar

0.33

 antecedentes

0.33

 fingerprint

0.33

 background

0.33

 badania

0.33

resher

0.31

 krwi

0.31

Activations Density 0.007%

terms related to background checks and criminal history records

background checks and clearance

No Comments

No Known Activations