INDEX

Explanations

introductory phrases that signal important information or requests

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

introducing a topic

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



-0.49

 betweenstory

-0.47

ChrTalk

-0.46

SqlClient

-0.45

GraphicsUnit

-0.45

kuuta

-0.43

thâu

-0.41

 Italijani

-0.40

odeon

-0.40

rzost

-0.39

POSITIVE LOGITS

 invokingState

0.54

MemoryWarning

0.47

 GenerationType

0.45

Weiterlesen

0.43

//

0.42

OrBuilder

0.41

AutoScale

0.41

yves

0.40

ptăm

0.40

 Comprehensive

0.39

Activations Density 0.002%

introductory phrases that signal important information or requests

introducing a topic

No Comments

No Known Activations