INDEX

Explanations

instances of technical or programming-related concepts and keywords

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

realise data cases focus concentration signature bigger

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

bootstrapcdn

-0.66

 مشين

-0.65

帖最后由

-0.59

脚注の使い方

-0.59

StructEnd

-0.59

تقاوى

-0.58

MemoryWarning

-0.57

rungsseite

-0.56

 Normdatei

-0.55

 Мексичка

-0.55

POSITIVE LOGITS

 metade

0.36

rev

0.34

ulosa

0.34

Surprisingly

0.34

insics

0.34

首

0.33

 noiva

0.33

 dibanding

0.33

BuildConfig

0.33

 balik

0.33

Activations Density 0.098%

instances of technical or programming-related concepts and keywords

realise data cases focus concentration signature bigger

No Comments

No Known Activations