INDEX

Explanations

references to specific tasks, samples, and data handling within practical contexts

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

protected class

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

Ƚ

-0.48

GEBURTSDATUM

-0.47

 AssemblyTitle

-0.47

principalColumn

-0.46

RTEX

-0.45

 PeEnEo

-0.45

.*")]

-0.45

 MainAxisSize

-0.44

 Taktlose

-0.44

CodeAttribute

-0.43

POSITIVE LOGITS

 dehydrated

0.32

 Вер

0.32

HtmlAttribute

0.31

樣的

0.31

 Marys

0.31

 Carbo

0.30

EClass

0.30

 Pico

0.30

getJSONArray

0.30

getStringExtra

0.29

Activations Density 0.193%

references to specific tasks, samples, and data handling within practical contexts

protected class

No Comments

No Known Activations