INDEX

Explanations

phrases that reflect clarity and evidence in reasoning or decision-making

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 linkovi

-0.68

RenderAtEndOf

-0.56

]")]

-0.53

 intStringLen

-0.52

 otomatig

-0.52

 ligiloj

-0.49

addContainerGap

-0.48

="@+

-0.48

Personendaten

-0.47

WebServlet

-0.47

POSITIVE LOGITS

 looking

2.25

Looking

2.02

 Looking

2.00

looking

1.90

 LOOKING

1.72

 olhando

1.62

LOOKING

1.48

 lookin

1.42

 examining

1.41

 mirando

1.26

Activations Density 0.853%

phrases that reflect clarity and evidence in reasoning or decision-making

No Comments

No Known Activations