INDEX

Explanations

references to personal and cultural backgrounds

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

background

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

SBATCH

-0.73



-0.71

 Jefus

-0.68

évaluateur

-0.65

UnitTesting

-0.64

 noDo

-0.62

 pleaſure

-0.61

 uſed

-0.59

 sánchez

-0.57

 myſelf

-0.57

POSITIVE LOGITS

 background

0.55

 Background

0.52

Background

0.48

 story

0.48

 backgrounds

0.45

background

0.44

 ویکی‌پدیای

0.43

 背景

0.43

 BACKGROUND

0.42

 heritage

0.41

Activations Density 0.016%

references to personal and cultural backgrounds

background

No Comments

No Known Activations