INDEX

Explanations

references to clickable links or call-to-action phrases in the document

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

"here for additional data file"

np_acts-logits-general · gemini-2.0-flash

click here for

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

SpringBootTest

-0.45

 AssemblyCulture

-0.42

Organisateur

-0.41

RSpec

-0.40

 calme

-0.36

Примечания

-0.36

 parís

-0.36

 milyen

-0.35

apimachinery

-0.35

initComponents

-0.35

POSITIVE LOGITS

 defaultstate

0.72

 button

0.63

 buttons

0.58

 Roskov

0.58

 link

0.57

 click

0.55

button

0.55

PositiveButton

0.54

 BUTTON

0.51

 clicked

0.49

Activations Density 0.157%

references to clickable links or call-to-action phrases in the document

"here for additional data file"

click here for

No Comments

No Known Activations