INDEX

Explanations

terms associated with data acquisition or results

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

crombie

-0.50

 comics

-0.47

 touch

-0.47

akei

-0.46

 ventilated

-0.46

 Perſ

-0.46

 facet

-0.45

erei

-0.45

estone

-0.44

continence

-0.44

POSITIVE LOGITS

 acquired

1.54

 obtained

1.45

acquired

1.42

obtained

1.30

 Acquired

1.24

 gained

1.23

Acquired

1.17

 Obtained

1.15

 zís

1.10

 purchased

1.03

Activations Density 0.201%

terms associated with data acquisition or results

No Comments

No Known Activations