INDEX

Explanations

references to specific protocols in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 calendar

-0.43

 Eisen

-0.42

 Engine

-0.41

 iron

-0.40

 Moyen

-0.39

 peaceful

-0.38

 Peaceful

-0.38

Engine

-0.38

 town

-0.37

 Eastern

-0.37

POSITIVE LOGITS

 protocol

0.72

 Protocol

0.69

Protocol

0.66

parsedMessage

0.65

 disambiguazione

0.65

 protocols

0.65

 Protocols

0.62

 PROTOCOL

0.58

protocol

0.57



0.56

Activations Density 0.202%

references to specific protocols in various contexts

No Comments

No Known Activations