INDEX

Explanations

phrases that indicate information sharing or providing clarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 Entstehung

-0.47

 Aspekte

-0.41

 Zustimmung

-0.40

 Bedürfn

-0.40

 Meinung

-0.40

 falschen

-0.40

 halben

-0.39

 Verhält

-0.39

 Notwendigkeit

-0.39

<eos>

-0.39

POSITIVE LOGITS

 kasarigan

1.20

niſſe

1.16

 pleaſure

1.16

 myſelf

1.14

 houſe

1.14

 ſind

1.13

 ſche

1.12

$_(

1.12

 orthodontic

1.10

ſelves

1.09

Activations Density 9.807%

phrases that indicate information sharing or providing clarity

No Comments

No Known Activations