INDEX

Explanations

conversational interactions and social engagement

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

alleri

-0.41

 Geller

-0.41

 Cardinale

-0.40

 Dignity

-0.38

ViewChild

-0.38

 Faul

-0.38

 convenience

-0.37

ขึ้น

-0.37

orene

-0.37

POSITIVE LOGITS

 conversation

0.77

 conversations

0.66

 conversación

0.63

 gesprek

0.63

 discussion

0.60

 chatting

0.60

conversation

0.58

 chats

0.56

 Conversation

0.56

 talking

0.56

Activations Density 0.292%

conversational interactions and social engagement

No Comments

No Known Activations