INDEX

Explanations

discussions of theoretical principles and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 című

-0.51

 devenus

-0.51

 regionales

-0.48

 privato

-0.48

 berätt

-0.47

xito

-0.47

 Utilizamos

-0.46

 semblables

-0.44

 manqué

-0.44

 særlig

-0.44

POSITIVE LOGITS

 theoretically

1.22

 Theore

1.03

Ideally

0.95

Theore

0.92

 theore

0.91

sätzlich

0.91

 nominally

0.91

 technically

0.91

Technically

0.90

 Ideally

0.89

Activations Density 0.547%

discussions of theoretical principles and conditions

No Comments

No Known Activations