INDEX

Explanations

references to computer programming constructs and code structure

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

code headers and definitions

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

majánló

-0.73

 miniaturka

-0.69

 Biôgrafia

-0.69

 Geſch

-0.68

 typelib

-0.67

[*]

-0.67

EndProject

-0.67

TypedDataSet

-0.65

elemField

-0.63

fromnode

-0.63

POSITIVE LOGITS

WithMany

0.38

0.37

0.35

0.34

0.33

0.32

0.31

0.29

Activations Density 0.005%

references to computer programming constructs and code structure

code headers and definitions

No Comments

No Known Activations