renorm_lreg_sweep | collingray Workspace – Weights & Biases

Skip to main content

Collingray's group workspace

Group: renorm_lreg_sweep

1-20

of 48

Tags

Notes

Author

State

Finished

Start time

April 12th, 2024 6:49:35 AM

Runtime

54m 45s

Tracked hours

54m 42s

Run path

collingray/multilayer_sae/nmegfy30

OS

Linux-6.7.8-arch1-1-x86_64-with-glibc2.39

Python version

3.11.4

Git repository

git clone https://github.com/collingray/autoencoders

Git state

git checkout -b "47-ML_R1.0e-03_rtlog_rs1_LR=5.0e-04" 4cb847f832900a6a1259b824ec2632a3cece8fad

Command

<python with no main file> -f /home/collin/.local/share/jupyter/runtime/kernel-e30dbe2c-2335-402f-9079-66d6983bea7f.json

System Hardware

CPU count	6
Logical CPU count	12
GPU count	1
GPU type	NVIDIA GeForce RTX 4090

W&B CLI Version

0.16.4

Group

renorm_lreg_sweep

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 7 keys
- act_renorm_scale:
  1
- act_renorm_type:
  "log"
- beta1:
  0
- beta2:
  0.99
- lambda_reg:
  0.0010000000474974513
- lr:
  0.0005
- warmup_percent:
  0.05

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 41 keys
- avg_fvu:
  0.55859375
- avg_l0:
  1,475.0821520618556
- ▶
  feature_density.bins:[] 65 items
- ▶
  feature_density.values:[] 64 items
- layer_0.l1:
  10.25
- layer_0.loss:
  0.01312255859375
- layer_0.mse:
  0.00286865234375
- layer_1.l1:
  20.5
- layer_1.loss:
  0.0240478515625
- layer_1.mse:
  0.0035858154296875
- layer_2.l1:
  17.75
- layer_2.loss:
  0.0216064453125
- layer_2.mse:
  0.0038604736328125
- layer_3.l1:
  18.75
- layer_3.loss:
  0.022705078125
- layer_3.mse:
  0.00390625
- layer_4.l1:
  19
- layer_4.loss:
  0.02392578125
- layer_4.mse:
  0.004852294921875
- layer_5.l1:
  17.375
- layer_5.loss:
  0.0230712890625
- layer_5.mse:
  0.0057373046875
- layer_6.l1:
  16.5
- layer_6.loss:
  0.022216796875
- layer_6.mse:
  0.005767822265625
- layer_7.l1:
  15.375
- layer_7.loss:
  0.0201416015625
- layer_7.mse:
  0.0047607421875
- layer_8.l1:
  14.5
- layer_8.loss:
  0.0191650390625
- layer_8.mse:
  0.004669189453125
- layer_9.l1:
  13.5
- layer_9.loss:
  0.0169677734375
- layer_9.mse:
  0.0034942626953125
- layer_10.l1:
  9.5
- layer_10.loss:
  0.0120849609375
- layer_10.mse:
  0.0025787353515625
- layer_11.l1:
  8.875
- layer_11.loss:
  0.01129150390625
- layer_11.mse:
  0.0024566650390625
- lr:
  0.00000000200057335412

This run produced these artifacts as outputs. Total: 1. Learn more

run-nmegfy30-history:v0

Loading...