INDEX

Explanations

programming-related terms and syntax structures

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

user input

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

protobuf

-0.40

 broader

-0.37

far

-0.37

 Nicolai

-0.37

 noexcept

-0.37

TableBody

-0.36

embols

-0.35

TypedValue

-0.35

 embrace

-0.35

 Arhivirano

-0.34

POSITIVE LOGITS

 propOrder

0.72

 initComponents

0.66

Personendaten

0.63

AccessorTable

0.59

IntoConstraints

0.59

sizeCache

0.59

 ujednoznacz

0.57

StoryboardSegue

0.56

ArgsConstructor

0.55

UpInside

0.55

Activations Density 0.002%

programming-related terms and syntax structures

user input

No Comments

No Known Activations