INDEX

Explanations

references to authentic research and experiential learning techniques

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

authentic experiential learning

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 geschlagen

-0.45

 drauf

-0.40

 sonriendo

-0.39

 normalidad

-0.39

 Daß

-0.39

 betrug

-0.38

 Schofield

-0.38

 rebaj

-0.38

utuhkan

-0.38

 Wikiseite

-0.37

POSITIVE LOGITS

Portale

0.59

 cherchés

0.55

AxisAlignment

0.51

ValueStyle

0.50

 CURLOPT

0.49

UserScript

0.48

DeleteBehavior

0.46

 experiential

0.46

-------

0.45

GeneratedMessage

0.45

Activations Density 0.018%

references to authentic research and experiential learning techniques

authentic experiential learning

No Comments

No Known Activations