INDEX

Explanations

symbols and structural elements commonly found in programming or mathematical notation

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

found, case, observation

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

 protoimpl

-0.69

 initComponents

-0.57

inSlope

-0.54

bootstrapcdn

-0.52

 rhestr

-0.51

IntoConstraints

-0.50

 Biôgrafia

-0.49

ellido

-0.48

 Audiodateien

-0.45

Попис

-0.45

POSITIVE LOGITS

Demografia

0.40

 مباش

0.39

évaluateur

0.37

 immediate

0.34

 Schatz

0.33

 inmediato

0.33

 geçen

0.31

 Scout

0.31

 yıld

0.30

 esternos

0.30

Activations Density 0.008%

symbols and structural elements commonly found in programming or mathematical notation

found, case, observation

No Comments

No Known Activations