INDEX

Explanations

references to newness or recently introduced concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 ब्रेकडाउन

-0.47

 Muerte

-0.42

UIControlState

-0.42

 redemption

-0.41

 Saludos

-0.41

 ويكيميديا

-0.40

principalTable

-0.40

plasia

-0.40

 blog

-0.40

 škoda

-0.40

POSITIVE LOGITS

 autorytatywna

0.49

PageRoute

0.42

Viungo

0.42

 realis

0.42

 actionTypes

0.41

QMetaType

0.41

Greater

0.40

ActionPerformed

0.40

 realise

0.39

 Conley

0.39

Activations Density 0.062%

references to newness or recently introduced concepts

No Comments

No Known Activations