INDEX

Explanations

references to qualifications and competency

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

familiar, qualified, values, behavior

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

 informée

-0.48

GEBURTSDATUM

-0.47

الحياه

-0.47

GraphicsUnit

-0.43

richTextPanel

-0.43

 تانيه

-0.42

 MainAxisSize

-0.42

 manguera

-0.41

 adecu

-0.41

complexContent

-0.40

POSITIVE LOGITS

 typelib

0.52

EndContext

0.44

 Clayton

0.43

 Butterfield

0.42

 Hick

0.41



0.40

 gradient

0.39

webElementXpaths

0.38

 Shir

0.38

pra

0.38

Activations Density 0.067%

references to qualifications and competency

familiar, qualified, values, behavior

No Comments

No Known Activations