INDEX

Explanations

numerical data and related terminology in scientific contexts

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

numerical simulations

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

égories

-0.56

findpost

-0.54

 pleaſure

-0.52

Personendaten

-0.52

 éter

-0.51

EnableWeb

-0.48

 itſelf

-0.48

 FetchType

-0.48

Становништво

-0.48

wiſe

-0.47

POSITIVE LOGITS

 simulations

0.42

 ['./

0.36

 Realität

0.35

AccessorTable

0.34

 MonoBehaviour

0.33

 computations

0.33

kozások

0.33

 computational

0.32

bux

0.32

 experiments

0.32

Activations Density 0.168%

numerical data and related terminology in scientific contexts

numerical simulations

No Comments

No Known Activations