INDEX

Explanations

phrases indicating wholeness or completeness in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

pe

-0.47

del

-0.45

čník

-0.45

laszt

-0.45

ron

-0.43

fo

-0.42

ster

-0.41

po

-0.41

sim

-0.40

POSITIVE LOGITS

 whole

2.48

whole

2.33

Whole

2.08

 Whole

1.95

 WHOLE

1.92

 entirety

1.58

 entire

1.44

 Ganzen

1.32

 totality

1.32

 geheel

1.32

Activations Density 0.166%

phrases indicating wholeness or completeness in context

No Comments

No Known Activations