INDEX

Explanations

technical terms and code syntax in programming contexts

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

Code file paths/namespaces

np_acts-logits-general · gemini-2.0-flash

file paths and module names

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

expandindo

-0.88

IntoConstraints

-0.86

GEBURTSDATUM

-0.79

 nakalista

-0.75

 BoxFit

-0.72

Personendaten

-0.68

 препратки

-0.67

RTLD

-0.66

 &___

-0.66

المكان

-0.66

POSITIVE LOGITS

 yür

0.34

 franquicia

0.33

 olvido

0.33

 oídos

0.32

 Nación

0.32

Berikut

0.31

ริง

0.31

 désert

0.30

บริ

0.30

 programu

0.30

Activations Density 0.107%

technical terms and code syntax in programming contexts

Code file paths/namespaces

file paths and module names

No Comments

No Known Activations