INDEX

Explanations

trends and data related to health metrics and outcomes

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

comparing higher or lower

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

 ModelExpression

-0.49

unhofer

-0.42

 beschik

-0.41

تقاوى

-0.40

 reformation

-0.40

intura

-0.39

zysta

-0.39

RefNanny

-0.38

 Dismiss

-0.37

uset

-0.37

POSITIVE LOGITS

 worsen

0.57

 worsened

0.56

 worse

0.56

 peor

0.54

Worse

0.51

 peggio

0.49

ViewFeatures

0.48

 worst

0.47

 worsening

0.47

 exacerbate

0.47

Activations Density 0.198%

trends and data related to health metrics and outcomes

comparing higher or lower

No Comments

No Known Activations