INDEX

Explanations

terms related to etiquette and respectful behavior in social contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 rhestr

-0.61

sidemargin

-0.57

cloudf

-0.47

landen

-0.47

InitVars

-0.47

 protoimpl

-0.46

 صوتيه

-0.46

chapper

-0.46

 ventricle

-0.46

czej

-0.46

POSITIVE LOGITS

 etiquette

0.77

 behavior

0.74

 manners

0.72

 Verhalten

0.69

 behaviors

0.67

 Etiquette

0.64

헌

0.62

 behaviour

0.61

cipline

0.60

 behave

0.60

Activations Density 0.261%

terms related to etiquette and respectful behavior in social contexts

No Comments

No Known Activations