INDEX

Explanations

references to sections in documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

alakip

-0.65

 ilustracja

-0.62

 autón

-0.61

 geïsole

-0.59

 ilustracji

-0.59

 Infór

-0.59

 illustrationer

-0.59

 leyendas

-0.58

 illustrazione

-0.57

 dezelve

-0.57

POSITIVE LOGITS

Sec

0.70

sec

0.67

Sec

0.66

sec

0.60

ts

0.54

lee

0.53

ter

0.51

SEC

0.49

tele

0.48

 tele

0.48

Activations Density 0.271%

references to sections in documents

No Comments

No Known Activations