INDEX

Explanations

references to new applications or software developments

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

 usually

-0.51

 Usually

-0.48

 suelen

-0.48

 suele

-0.48

 colis

-0.46

 brukar

-0.45

 habituales

-0.44

 çoğu

-0.44

HasColumnName

-0.44

 tocado

-0.43

POSITIVE LOGITS

 inaugural

0.57

新たに

0.51

INTRODU

0.50

new

0.50

 launch

0.49

Launching

0.47

új

0.46

 newly

0.46

新しく

0.46

RenderAtEndOf

0.46

Activations Density 0.433%

references to new applications or software developments

No Comments

No Known Activations