INDEX

Explanations

references to plainness or simplicity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

neſs

-0.60

SPJ

-0.59

 pleaſure

-0.58

 Chrift

-0.57

 Kette

-0.56

eien

-0.56

 Résultats

-0.55

 Monfieur

-0.55

 Jefus

-0.54

 houſe

-0.54

POSITIVE LOGITS

Feature

0.88

 feature

0.86

feature

0.77

 Feature

0.75

intios

0.73

 features

0.72

din

0.67

 Biôgrafia

0.67

bbat

0.64

renderItem

0.63

Activations Density 0.094%

references to plainness or simplicity

No Comments

No Known Activations