INDEX

Explanations

technical outputs and performance metrics related to model loading and processing times

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 39-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_39/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.39.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

FirstResponder

-0.41

 sprá

-0.39

ravel

-0.39

ocats

-0.38

 Pelle

-0.38

 felf

-0.38

 assise

-0.37

ObjectContext

-0.36

 Turquía

-0.36

forChild

-0.36

POSITIVE LOGITS

 train

0.75

 epoch

0.73

 Train

0.65

 trained

0.64

 epochs

0.63

 model

0.63

train

0.62

 checkpoint

0.61

Train

0.61

 trainer

0.60

Activations Density 0.193%

technical outputs and performance metrics related to model loading and processing times

No Comments

No Known Activations