INDEX

Explanations

technical terms related to electronics or circuits

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

Actions of turning something on/off

np_acts-logits-general · gemini-2.0-flash

turned on or off

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 '\\;'

-0.75

ⓧ

-0.60

 CURIAM

-0.60

 Meille

-0.58

 AspNetCore

-0.57

thâu

-0.56

bootstrapcdn

-0.56

verwijspagina

-0.55

 propOrder

-0.54

MergeFrom

-0.53

POSITIVE LOGITS

 turned

1.17

 Turned

1.09

 turn

1.09

turned

1.02

Turn

1.02

Turned

1.02

 Turn

1.00

 turning

0.99

 power

0.98

turn

0.97

Activations Density 0.292%

technical terms related to electronics or circuits

Actions of turning something on/off

turned on or off

No Comments

No Known Activations