INDEX

Explanations

references to modifications and versions of concepts or items

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

followed by a phrase

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

 protoimpl

-0.63

 utafitiHapana

-0.60

Hochspringen

-0.59

 arşivlendi

-0.56

:✨

-0.53

 الحره

-0.50

Referències

-0.50

+#+

-0.49

İstinadlar

-0.49

 mijne

-0.49

POSITIVE LOGITS

sze

0.37

pter

0.36

another

0.35

olge

0.34

 another

0.33

low

0.33

 supplement

0.32

 ComVisible

0.32

oso

0.32

تر

0.32

Activations Density 0.056%

references to modifications and versions of concepts or items

followed by a phrase

No Comments

No Known Activations