INDEX

Explanations

specific game-related terms and instructions

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

Game hints or instructions

np_acts-logits-general · gemini-2.0-flash

level secrets and objectives

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 utafitiHapana

-0.95

tvguidetime

-0.84

DebuggerNonUser

-0.73

TypedDataSet

-0.71

الحياه

-0.69

 collectivités

-0.63

AutoresizingMask

-0.62

 المعيارى

-0.62

ftagPool

-0.61

PyExc

-0.61

POSITIVE LOGITS

 interac

0.68

 unlocked

0.64

 unlock

0.64

 quest

0.61

 nearby

0.60

 upstairs

0.59

 checkpoint

0.58

 respawn

0.57

 mission

0.56

 hidden

0.56

Activations Density 0.335%

specific game-related terms and instructions

Game hints or instructions

level secrets and objectives

No Comments

No Known Activations