INDEX

Explanations

items associated with weapons, specifically knives and swords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Lähteet

-0.72

hoots

-0.65

żesz

-0.64

舺

-0.63

 Wallflower

-0.60

 виправивши

-0.57

Cyfeiriadau

-0.56

UnknownFieldSet

-0.56

jaya

-0.55

 gustó

-0.55

POSITIVE LOGITS

 blade

1.26

 knife

1.19

 blades

1.07

 knives

1.04

 Blade

1.02

Blade

0.98

blade

0.96

 Knife

0.94

Knife

0.93

 Blades

0.90

Activations Density 0.091%

items associated with weapons, specifically knives and swords

No Comments

No Known Activations