INDEX

Explanations

programming syntax and specific characters used in coding

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

syntax and mathematical notation

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

parsedMessage

-0.88

endphp

-0.82

 queſta

-0.80

<pad>

-0.80

<unused79>

-0.79

<unused74>

-0.79

<unused16>

-0.79

<unused8>

-0.79

<unused3>

-0.79

[@BOS@]

-0.79

POSITIVE LOGITS

 cheminée

0.44

 Cataluña

0.39

 fuite

0.36

 Ebenso

0.35

 Spitze

0.35

 Velasco

0.33

 alimentação

0.32

•

0.32

 okazji

0.32

 prensa

0.31

Activations Density 0.025%

programming syntax and specific characters used in coding

syntax and mathematical notation

No Comments

No Known Activations