INDEX

Explanations

programming constructs related to variable assignments and control structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 ویکی‌پدی

-0.61

 MenuView

-0.59

awaiter

-0.58

WriteTagHelper

-0.57

GEBURTSDATUM

-0.57

 CURIAM

-0.57

 dAtA

-0.57

PreferredItem

-0.56

toBeTruthy

-0.55

MarshalTo

-0.55

POSITIVE LOGITS

 separately

0.37

 both

0.34

Ծանոթ

0.33

bigskip

0.32

 elsewhere

0.31

 liars

0.31

 simultaneously

0.30

 couple

0.30

 unfortunately

0.30

 stag

0.30

Activations Density 0.433%

programming constructs related to variable assignments and control structures

No Comments

No Known Activations