INDEX

Explanations

elements related to participation in events and locations

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

walking outdoors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

出版年

-0.46

 AssemblyProduct

-0.43

новништво

-0.41

AutoScaleMode

-0.41

Personendaten

-0.41

-0.40

 Preferencias

-0.38

OrUpdate

-0.38

StoredProcedure

-0.36

最快更新

-0.36

POSITIVE LOGITS

 walking

0.57

 scattered

0.56

Walking

0.55

 Walking

0.54

walking

0.52

 WALK

0.52

 walk

0.49

 streets

0.49

 walks

0.48

 outdoors

0.47

Activations Density 0.063%

elements related to participation in events and locations

walking outdoors

No Comments

No Known Activations