INDEX

Explanations

references to physical locations or positions related to relative depth or bottom placement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 CreateTagHelper

-0.59

 StringTokenizer

-0.54

 Twig

-0.52

prepareStatement

-0.51

canf

-0.49

Twig

-0.48

zzleHttp

-0.48

XMLHttpRequest

-0.48

createStatement

-0.48

eid

-0.47

POSITIVE LOGITS

 bottom

0.96

bottom

0.94

 down

0.87

down

0.85

 BOTTOM

0.85

 below

0.84

 abajo

0.83

Bottom

0.80

 downward

0.79

Down

0.79

Activations Density 0.659%

references to physical locations or positions related to relative depth or bottom placement

No Comments

No Known Activations