INDEX

Explanations

sections of text without any content, indicating a lack of relevant information

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

Following questions or acknowledgements

np_acts-logits-general · gemini-2.0-flash

hope or wish

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

Referencie

-0.61

Билгалдахарш

-0.57

 resourceCulture

-0.55

 ogóle

-0.52

 usus

-0.51

bleshooting

-0.51

 Drapeau

-0.51

 économies

-0.50

Dono

-0.50

Elő

-0.49

POSITIVE LOGITS

 hope

0.88

 hopes

0.73

hope

0.72

Hope

0.70

 Hope

0.67

 apolog

0.62

 wish

0.61

 hopeful

0.58

ItemBackground

0.58

 hoffen

0.56

Activations Density 0.175%

sections of text without any content, indicating a lack of relevant information

Following questions or acknowledgements

hope or wish

No Comments

No Known Activations