INDEX

Explanations

references to algorithms and programming concepts

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

algorithm paths

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

ReusableCell

-0.59

MessageTagHelper

-0.54

tagHelperRunner

-0.51

 estekak

-0.50

newArrayList

-0.48

exitRule

-0.44

RenderAtEndOf

-0.44

PerformLayout

-0.43

fromnode

-0.42

IActionResult

-0.41

POSITIVE LOGITS

 compactness

0.52

andale

0.51

Viited

0.47

 cuticle

0.46

 disponibilités

0.45

Induction

0.44

 compact

0.44

 anisotropy

0.42

Armando

0.42

 floors

0.42

Activations Density 0.000%

references to algorithms and programming concepts

algorithm paths

No Comments

No Known Activations