INDEX

Explanations

terms related to programming configurations and library management

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

build configuration and conversational text

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

 betweenstory

-0.43

 actionMode

-0.42

MessageOf

-0.42

+#+#

-0.41

ghalaya

-0.39

oredCriteria

-0.39

 ویکی‌پدی

-0.38

GraphicsUnit

-0.38

évaluateur

-0.38

POSITIVE LOGITS

 nahilalakip

0.50

 NgModule

0.49

BibitemShut

0.49

NgModule

0.48

 Sight

0.46

exitRule

0.46

vitis

0.45

rungsseite

0.44

ódio

0.44

 Verſ

0.44

Activations Density 0.012%

terms related to programming configurations and library management

build configuration and conversational text

No Comments

No Known Activations