INDEX

Explanations

references to top or best performers in various fields

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

best and top people

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

Negative Logits

RenderAtEndOf

-0.40

Gegen

-0.40

embatan

-0.36

告

-0.36

ècie

-0.35

WebControls

-0.33

 Kayak

-0.33

Bag

-0.32

bag

-0.32

greSQL

-0.32

POSITIVE LOGITS

 AssemblyCompany

0.63

 للاسماء

0.61

 beginnetje

0.60

CppMethod

0.59

excellence

0.58

RectangleBorder

0.57

 excellence

0.56

 '\\;'

0.56

Excellence

0.56

 лучшие

0.56

Activations Density 0.021%

references to top or best performers in various fields

best and top people

No Known Activations