INDEX

Explanations

elements related to emotional or dramatic experiences

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

 himſelf

-1.21

 myſelf

-1.11

 themſelves

-1.10

 itſelf

-1.07

Efq

-1.07

 Monfieur

-1.06

 purpoſe

-1.03

 Houſe

-1.03

Eſ

-1.02

 houſe

-1.01

POSITIVE LOGITS

ra

0.52

rad

0.44

pil

0.43

かた

0.41

BOD

0.40

cr

0.40

 cellspacing

0.39

 Perhaps

0.39

mell

0.39

xp

0.38

Activations Density 0.407%

elements related to emotional or dramatic experiences

No Comments

No Known Activations