INDEX

Explanations

references to variables and their management in a programming context

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

gradients and tensor shapes

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

 AssemblyTitle

-0.56

 AppCompatTheme

-0.53

IUrlHelper

-0.44

UnitTesting

-0.44

 مرئيه

-0.41

 navide

-0.40

 šte

-0.37

 hombro

-0.36

cèse

-0.36

ⓧ

-0.36

POSITIVE LOGITS

viewDidLoad

0.48

 Wiktionnaire

0.41

clusion

0.41

 ویکی‌پدیا

0.41

cipline

0.41

OCCURRED

0.41

redients

0.40

colleg

0.40

 vesting

0.40

bat

0.38

Activations Density 0.042%

references to variables and their management in a programming context

gradients and tensor shapes

No Comments

No Known Activations