INDEX

Explanations

actions related to flipping or rolling over

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

новниш

-0.47

 poveznice

-0.46

Đi

-0.45

 Popis

-0.45

どうしても

-0.45

 plads

-0.44

 oublié

-0.44

 maaf

-0.44

்கு

-0.43

Po

-0.43

POSITIVE LOGITS

 flipped

1.41

 flips

1.41

 flipping

1.39

 flip

1.36

flip

1.26

Flip

1.24

 Flip

1.19

 overturn

1.15

turning

1.14

 Turning

1.13

Activations Density 0.217%

actions related to flipping or rolling over

No Comments

No Known Activations