INDEX

Explanations

structured inquiries or criteria related to evaluating information and claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

adaptiveStyles

-0.71

PreferredItem

-0.70

 autorytatywna

-0.70

migrationBuilder

-0.66

Personendaten

-0.65

sizeCache

-0.63

MessageTagHelper

-0.63

Portail

-0.63

 ſei

-0.62

 disambiguazione

-0.62

POSITIVE LOGITS

0.98

0.95

0.94

0.90

0.89

0.85

0.84

0.79

Activations Density 1.292%

structured inquiries or criteria related to evaluating information and claims

No Comments

No Known Activations