INDEX

Explanations

questions about processes and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ContentAlignment

-0.47

 kysy

-0.42

 sério

-0.41

 dezelve

-0.41

 brancas

-0.41

 econômica

-0.40

 mitään

-0.39

 سكانية

-0.39

 kaupung

-0.39

 kirja

-0.38

POSITIVE LOGITS

How

0.87

How

0.85

how

0.84

how

0.79

Cómo

0.79

 cómo

0.77

HOW

0.75

howto

0.75

 如何

0.71

HOW

0.69

Activations Density 0.242%

questions about processes and experiences

No Comments

No Known Activations