INDEX

Explanations

references to identifiable user information and privacy protection

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

identifiable information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 writeTo

-0.41

seiti

-0.40

Proposed

-0.40

 resourceCulture

-0.38

σκε

-0.38

proposed

-0.37

iocene

-0.36

therapeutic

-0.36

Developmental

-0.36

튕

-0.36

POSITIVE LOGITS

 identifiable

0.94

 Identi

0.94

 identific

0.92

 identifica

0.92

 identification

0.90

 identi

0.87

identi

0.86

Identi

0.86

 identif

0.85

 identifying

0.84

Activations Density 0.020%

references to identifiable user information and privacy protection

identifiable information

No Comments

No Known Activations