INDEX

Explanations

phrases indicating past actions 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 @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_9/width_16k/average_l0_100

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

VYMaps

-0.57

WriteTagHelper

-0.48

 jsPsych

-0.44

resave

-0.42

approche

-0.40

Réalisation

-0.40

 Италијани

-0.39

lectricité

-0.39

뀐

-0.38

Unset

-0.38

POSITIVE LOGITS

 ooit

0.76

 once

0.70

 kiedyś

0.69

 pernah

0.69

曾經

0.67

 einst

0.67

 egyszer

0.66

once

0.64

 Once

0.57

 someday

0.57

Activations Density 0.248%

phrases indicating past actions and experiences

No Comments

No Known Activations