INDEX

Explanations

references to skills and expertise

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

skills, knowledge, and authority

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

 '\\;'

-0.56

 comuniques

-0.55

ViewFeatures

-0.54

AddTagHelper

-0.54

 صوتيه

-0.53

 Италијани

-0.52

ariConfig

-0.50

ipients

-0.50

/\.

-0.50

 فريبيس

-0.49

POSITIVE LOGITS

 expertise

0.46

 experto

0.44

 esfuerzo

0.40

 skills

0.39

 esfuer

0.39

 humilde

0.38

 aprovecha

0.37

wissen

0.36

 applied

0.36

NameInMap

0.36

Activations Density 0.026%

references to skills and expertise

skills, knowledge, and authority

No Comments

No Known Activations