INDEX

Explanations

tokens related to classification or categorization in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 nahilalakip

-0.44

 CreateTagHelper

-0.43

 reciente

-0.41

 recently

-0.40

 faute

-0.39

Према

-0.38

 recent

-0.38

 Recently

-0.38

ilies

-0.36

 récemment

-0.36

POSITIVE LOGITS

TypedDataSet

0.63

 decades

0.61

 kiedyś

0.57

 yıllar

0.56

昔

0.56

当年

0.56

かつて

0.55

 crossorigin

0.54

AsUp

0.54

 damals

0.54

Activations Density 0.987%

tokens related to classification or categorization in various contexts

No Comments

No Known Activations