INDEX

Explanations

interrogative words and phrases, indicating a focus on questions and quests for information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 <<<<<<<<<<<<<<

-0.94

 BoxDecoration

-0.94

URLException

-0.90

 lenker

-0.89

MemoryWarning

-0.88

homonymie

-0.88

Aiheesta

-0.83

XmlAccessorType

-0.82

etheless

-0.80

μως

-0.80

POSITIVE LOGITS

What

0.58

Is

0.56

My

0.53

De

0.51

Per

0.51

By

0.50

Where

0.50

Do

0.50

How

0.50

my

0.49

Activations Density 0.521%

interrogative words and phrases, indicating a focus on questions and quests for information

No Comments

No Known Activations