INDEX

Explanations

the phrase "du," likely focusing on patterns or repetitions in language usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 ainfi

-0.96

 iſt

-0.94

Eſ

-0.91

 Monfieur

-0.91

 myſelf

-0.90

 faſt

-0.85

 itſelf

-0.85

 purpoſe

-0.84

 Diſ

-0.83

 ſtate

-0.83

POSITIVE LOGITS

du

3.09

Du

1.48

del

1.19

Du

1.14

des

1.02

le

0.95

DU

0.85

 delle

0.77

Le

0.72

 della

0.72

Activations Density 0.076%

the phrase "du," likely focusing on patterns or repetitions in language usage

No Comments

No Known Activations