INDEX

Explanations

elements related to educational content and learning activities

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

questions method instruction clues

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

LookAnd

-0.58

PasswordEncoder

-0.41

bkz

-0.37

getMenuInflater

-0.36

 JpaRepository

-0.36

 ProtoMessage

-0.36

 ©️

-0.35

 AssemblyTitle

-0.34

 محفوظة

-0.34

WebpackPlugin

-0.34

POSITIVE LOGITS

OGND

0.51

 Jefus

0.50

 мәкал

0.48

colt

0.48

 revision

0.47

 وصلة

0.47

 שוליים

0.47

ſelf

0.45

сылкі

0.45

Fä

0.45

Activations Density 0.005%

elements related to educational content and learning activities

questions method instruction clues

No Comments

No Known Activations