INDEX

Explanations

patterns related to data handling and organization in documents

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

respond to

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

 estekak

-0.54

exitRule

-0.53

TypedDataSet

-0.52

NamedQueries

-0.45

ViewFeatures

-0.45

 <=",

-0.44

StatusBadRequest

-0.42

 BoxFit

-0.41

 fieldNum

-0.41

GTCX

-0.40

POSITIVE LOGITS

Ligações

0.44

gameserver

0.43

 nakalista

0.42

teel

0.41

eqn

0.38

Had

0.36

Slf

0.36

 fühlte

0.35

phazard

0.35

peł

0.34

Activations Density 0.188%

patterns related to data handling and organization in documents

respond to

No Comments

No Known Activations