INDEX

Explanations

terms related to automated processes or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

contentLoaded

-0.93

 Anfitrión

-0.91

 StringField

-0.90

URLException

-0.88

jspx

-0.88

Personensuche

-0.86

adpleegd

-0.85

 useAppContext

-0.84

IVEREF

-0.83

Kraken

-0.83

POSITIVE LOGITS

 Auto

1.74

 AUTO

1.73

 auto

1.67

Auto

1.61

auto

1.60

AUTO

1.58

setAuto

1.29

 autos

1.13

 Autos

1.09

autos

1.06

Activations Density 0.057%

terms related to automated processes or systems

No Comments

No Known Activations