INDEX

Explanations

terms related to specific technical components and their characteristics in various contexts

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

preceding technical terms

np_acts-logits-general · gemini-2.0-flash

layered structure, chair, lens, directional signature, array

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 typelib

-0.46

 AssemblyTitle

-0.42

 textStatus

-0.40

 stripper

-0.38

 protoimpl

-0.38

[*]

-0.37

kembang

-0.37

strich

-0.37

 cister

-0.37

 comprising

-0.37

POSITIVE LOGITS

Tazama

0.59

 pinulongan

0.57

 ویکی‌پدی

0.52

الحياه

0.47

 الحره

0.46

 Мексичка

0.44

istoitu

0.43

 Reſ

0.42

 AttributeSet

0.42

 AssemblyCompany

0.41

Activations Density 1.368%

terms related to specific technical components and their characteristics in various contexts

preceding technical terms

layered structure, chair, lens, directional signature, array

No Comments

No Known Activations