INDEX

Explanations

technical jargon and specialized vocabulary related to scientific contexts and research

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

followed by "act" or "acts"

np_acts-logits-general · gemini-2.0-flash

describing purpose or function

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

enumi

-0.76

UnsafeEnabled

-0.55

تفصیلات

-0.52

 Requires

-0.51

 متعلقه

-0.51

Requires

-0.51

Composable

-0.50

 nameof

-0.49

に進

-0.48

mogorov

-0.47

POSITIVE LOGITS

act

2.42

 acts

2.29

 acting

2.23

 acted

2.23

 serve

2.14

 serves

2.03

act

1.95

Act

1.91

Act

1.88

 serving

1.82

Activations Density 0.820%

technical jargon and specialized vocabulary related to scientific contexts and research

followed by "act" or "acts"

describing purpose or function

No Comments

No Known Activations