INDEX

Explanations

words and phrases related to training and practice

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

 referenties

-0.47

MigrationBuilder

-0.45

 Numerade

-0.45

 jure

-0.45

 مشين

-0.43

ğraf

-0.43

OrNil

-0.43

ytale

-0.43

 unie

-0.41

 IconData

-0.41

POSITIVE LOGITS

 experience

0.66

 experiences

0.62

 experiencia

0.60

Experience

0.59

 Experience

0.57

 erfaring

0.56

 EXPERIENCE

0.55

experience

0.54

 practice

0.54

 pengalaman

0.54

Activations Density 0.245%

words and phrases related to training and practice

No Comments

No Known Activations