INDEX

Explanations

expressions of preference or choice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

</em>

-0.78

道

-0.68

zel

-0.63

-0.60

त्व

-0.59

Tom

-0.59

<em>

-0.58

<h3>

-0.58

</i>

-0.58

it

-0.58

POSITIVE LOGITS

 prefer

2.46

 preferred

2.42

 Prefer

2.42

 preference

2.34

prefer

2.31

 Preferred

2.26

 prefers

2.26

preferred

2.21

 Preference

2.18

 préfé

2.14

Activations Density 0.054%

expressions of preference or choice

No Comments

No Known Activations