INDEX

Explanations

terminology related to control and regulation, particularly in scientific or medical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

issau

-0.65

IBarButtonItem

-0.62

hoga

-0.62

apimachinery

-0.61

 resourceCulture

-0.60

ashqai

-0.59

Teb

-0.58

 Volatility

-0.58

 Wikimedijinoj

-0.57

AndEndTag

-0.57

POSITIVE LOGITS

 controlled

2.01

controlled

1.75

Controlled

1.71

 Controlled

1.69

 managed

1.64

 supervised

1.55

managed

1.54

 operated

1.50

 Managed

1.50

Managed

1.42

Activations Density 0.219%

terminology related to control and regulation, particularly in scientific or medical contexts

No Comments

No Known Activations