INDEX

Explanations

elements related to products and their details in a structured format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

Sitz

-0.35

 gebo

-0.35

RuleContext

-0.33

 alianza

-0.32

 costas

-0.32

 fondation

-0.31

 Charters

-0.31

 angele

-0.31

 abraço

-0.31

 instituições

-0.31

POSITIVE LOGITS

 retailing

0.72

 price

0.68

 sellers

0.67

 selling

0.66

 seller

0.61

 merchandising

0.61

 thiệu

0.60

 AssemblyCompany

0.60

featureID

0.60

 retailers

0.59

Activations Density 0.196%

elements related to products and their details in a structured format

No Comments

No Known Activations