INDEX

Explanations

actions related to movement and walking

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mather

-0.57

 extrême

-0.49

 jasa

-0.49

 Vierge

-0.48

 mercy

-0.48

 Needham

-0.48

Es

-0.48

 änd

-0.48

 bienvenida

-0.47

atoma

-0.47

POSITIVE LOGITS

 walking

1.62

 walk

1.58

 walked

1.52

 walks

1.50

Walking

1.48

 WALK

1.47

 Walking

1.47

walking

1.46

 Walk

1.43

walk

1.42

Activations Density 0.198%

actions related to movement and walking

No Comments

No Known Activations