INDEX

Explanations

specific characters or sequences related to programming or code structures

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

creva

-0.45

 debió

-0.45

må

-0.44

thâu

-0.44

 betere

-0.43

 necesariamente

-0.43

 devaient

-0.43

osocial

-0.42

 moeten

-0.42

 criança

-0.42

POSITIVE LOGITS

KommentareTeilen

0.75

0.69

FXML

0.64

principalColumn

0.61

 protoimpl

0.56

IBOutlet

0.56

 disambiguazione

0.56

elemField

0.54

SourceChecksum

0.54

asteroide

0.52

Activations Density 0.113%

specific characters or sequences related to programming or code structures

No Comments

No Known Activations