INDEX

Explanations

mentions of new items or developments across various contexts

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

new followed by a noun

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

 snippetHide

-0.54

hoga

-0.51

 PeEnEo

-0.49

HideFlags

-0.49

 pinulongan

-0.49

➯

-0.48

+#+

-0.47

 OFDb

-0.45

 chase

-0.45

 nonUne

-0.44

POSITIVE LOGITS

azgo

0.48

horabuena

0.40

føl

0.36

 løpet

0.35

 Grecs

0.35

󠁮

0.35

 förs

0.34

 newState

0.34

occasione

0.33

vább

0.33

Activations Density 0.012%

mentions of new items or developments across various contexts

new followed by a noun

No Comments

No Known Activations