collingray

Collingray's group workspace

Group: multilayer_pythia_32x_v4

1-16

of 16

Tags

Notes

Author

collingray

State

Finished

Start time

May 10th, 2024 2:53:00 AM

Runtime

49m 42s

Tracked hours

49m 39s

Run path

collingray/multilayer_sae/wekv32as

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

Python version

3.11.4

Command

<python with no main file> -f /home/collin/.local/share/jupyter/runtime/kernel-098de2af-2c9e-47c3-9e3c-672358f850c3.json

System Hardware

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

W&B CLI Version

0.16.1

Group

multilayer_pythia_32x_v4

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 12 keys
- act_renorm_scale:
  1
- act_renorm_type:
  "none"
- decoder_scale:
  true
- input_scale:
  false
- latent_p:
  1
- lr:
  0.001
- lr_decay_pct:
  0.2
- lr_warmup_pct:
  0
- β1:
  0.9
- β2:
  0.999
- λ:
  0.04641588777303696
- λ_warmup_pct:
  null

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 47 keys
- layer_0.avg_fvu:
  0.2784806191921234
- layer_0.avg_l0:
  101.12397003173828
- ▶
  layer_0.feature_density.bins:[] 65 items
- ▶
  layer_0.feature_density.values:[] 64 items
- layer_0.l1:
  0.0118408203125
- layer_0.loss:
  0.330078125
- layer_0.mse:
  0.330078125
- layer_1.avg_fvu:
  0.3081412613391876
- layer_1.avg_l0:
  236.1965484619141
- ▶
  layer_1.feature_density.bins:[] 65 items
- ▶
  layer_1.feature_density.values:[] 64 items
- layer_1.l1:
  5.53125
- layer_1.loss:
  0.62109375
- layer_1.mse:
  0.365234375
- layer_2.avg_fvu:
  0.23234181106090543
- layer_2.avg_l0:
  3,247.303955078125
- ▶
  layer_2.feature_density.bins:[] 65 items
- ▶
  layer_2.feature_density.values:[] 64 items
- layer_2.l1:
  80
- layer_2.loss:
  4
- layer_2.mse:
  0.275390625
- layer_3.avg_fvu:
  0.3345063030719757
- layer_3.avg_l0:
  261.0754699707031
- ▶
  layer_3.feature_density.bins:[] 65 items
- ▶
  layer_3.feature_density.values:[] 64 items
- layer_3.l1:
  1.546875
- layer_3.loss:
  0.46875
- layer_3.mse:
  0.396484375
- layer_4.avg_fvu:
  0.24387648701667783
- layer_4.avg_l0:
  1,638.9534912109375
- ▶
  layer_4.feature_density.bins:[] 65 items
- ▶
  layer_4.feature_density.values:[] 64 items
- layer_4.l1:
  53
- layer_4.loss:
  2.75
- layer_4.mse:
  0.2890625
- layer_5.avg_fvu:
  0.29001525044441223
- layer_5.avg_l0:
  2,396.950439453125
- ▶
  layer_5.feature_density.bins:[] 65 items
- ▶
  layer_5.feature_density.values:[] 64 items
- layer_5.l1:
  91.5
- layer_5.loss:
  4.59375
- layer_5.mse:
  0.34375
- lr:
  0.00001285120000000181
- total.avg_fvu:
  0.27765995264053345
- total.avg_l0:
  1,313.72119140625
- ▶
  total.feature_density.bins:[] 65 items
- ▶
  total.feature_density.values:[] 64 items

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

wandb-history

run-wekv32as-history:v0

Collingray's group workspace

multilayer_pythia_32x_v4

14: λ4.6e-02_LR1.0e-03_β₁0.9_β₂0.999