INDEX

Explanations

programming constructs and functions related to memory management and data handling

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

get/init/unique storage

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

Personendaten

-0.68

 otomatig

-0.66

OGND

-0.64

 مرئيه

-0.60

rungsseite

-0.60

 يتيمه

-0.60

Autoritní

-0.60

DockStyle

-0.59

󠁿

-0.59

 engraçadas

-0.58

POSITIVE LOGITS

Freshman

0.36

Sinopse

0.36

-------

0.36

NonNull

0.35

futbolista

0.35

DeltaTime

0.35

 Göteborg

0.34

Datuak

0.34

 geng

0.34

DISCLAIMER

0.34

Activations Density 0.147%

programming constructs and functions related to memory management and data handling

get/init/unique storage

No Comments

No Known Activations