INDEX

Explanations

keywords related to programming structure and authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 ménage

-0.33

 Había

-0.32

Wat

-0.32

Wat

-0.31

 masiva

-0.29

 čier

-0.29

 passed

-0.29

 recevrez

-0.29

wat

-0.28

 aikaa

-0.28

POSITIVE LOGITS

ſicht

0.64

:✨

0.63

ロウィン

0.63

 ſtate

0.63

ſelves

0.62

aarrggbb

0.58

ngdoc

0.58

 ftate

0.58

ſelf

0.57

ſent

0.57

Activations Density 0.131%

keywords related to programming structure and authority

No Comments

No Known Activations