INDEX

Explanations

concepts and terminology related to learning, understanding, and information processing

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

новниш

-0.75

 Monfieur

-0.61

 houſe

-0.60

 perfons

-0.60

AddTagHelper

-0.57

jectures

-0.57

WebVitals

-0.56

новништво

-0.56

ſelves

-0.55

omination

-0.54

POSITIVE LOGITS

 absorption

0.42

받

0.40

 mémor

0.39

 zrozum

0.35

 leçon

0.35

 absorb

0.34

 aufnehmen

0.33

 Lern

0.33

 absorbing

0.33

 learning

0.32

Activations Density 0.295%

concepts and terminology related to learning, understanding, and information processing

No Comments

No Known Activations