INDEX

Explanations

instances of the word "walk" and its variations in different contexts

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

 navideños

-0.83

 Darmstadt

-0.81

 firepower

-0.79

 Brenner

-0.79

 suprême

-0.76

 purpoſe

-0.74

 cielos

-0.74

 Anya

-0.73

 miatt

-0.73

 sacerdotes

-0.72

POSITIVE LOGITS

 walk

2.68

 Walk

2.50

walk

2.48

 WALK

2.48

 walks

2.47

Walk

2.40

 walking

2.36

 walked

2.33

 Walks

2.24

WALK

2.18

Activations Density 0.048%

instances of the word "walk" and its variations in different contexts

No Comments

No Known Activations