INDEX

Explanations

categories and classifications in a structured format

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

Code syntax or "Category:" labels

np_acts-logits-general · gemini-2.0-flash

both categories

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

 TextAppearance

-0.69

Diweddarwch

-0.64

 ErrIntOverflow

-0.55

RotationOrder

-0.55

OGND

-0.52

Наводи

-0.48

 للاسماء

-0.48

 InputDecoration

-0.47

WriteTagHelper

-0.46

)";

-0.46

POSITIVE LOGITS

 both

0.71

 begge

0.67

Both

0.65

 ambos

0.65

both

0.65

 Both

0.64

 beide

0.63

 båda

0.63

Kedua

0.61

 beiden

0.59

Activations Density 0.949%

categories and classifications in a structured format

Code syntax or "Category:" labels

both categories

No Comments

No Known Activations