Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
__pycache__/
ckpt/*.pt
2,777 changes: 5 additions & 2,772 deletions AgentMemorySystem.py

Large diffs are not rendered by default.

60 changes: 60 additions & 0 deletions ckpt/train_log.jsonl

Large diffs are not rendered by default.

71 changes: 71 additions & 0 deletions ckpt/train_stdout.log
Original file line number Diff line number Diff line change
@@ -0,0 +1,71 @@
[build] d_LLM=1536 L_mem=8 dampen=0.25
`torch_dtype` is deprecated! Use `dtype` instead!
Loading weights: 0%| | 0/338 [00:00<?, ?it/s]Loading weights: 100%|██████████| 338/338 [00:00<00:00, 11767.67it/s]
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
[neighbor cache] vocab_size=151936 > 60000, skip
[build] device=cpu tok_pad=<|endoftext|>
[build] params total=1,657,083,224 trainable=113,368,920
[build] memories stored: 11
[step 0 | 8.0s] tot=566.851 recon=4.205 sa=9.902 et=5.552 tsa=10.922 va=-0.000 cs=0.000
[step 1 | 6.8s] tot=108.189 recon=4.312 sa=9.655 et=5.246 tsa=10.819 va=-0.033 cs=0.024
[step 2 | 7.1s] tot=65.034 recon=4.937 sa=9.591 et=5.510 tsa=10.674 va=-0.069 cs=0.017
[step 3 | 6.8s] tot=66.606 recon=5.500 sa=9.480 et=6.235 tsa=10.755 va=-0.103 cs=0.089
[step 4 | 7.1s] tot=62.108 recon=3.728 sa=9.583 et=5.185 tsa=10.267 va=-0.131 cs=0.118
[step 5 | 6.3s] tot=109.068 recon=3.722 sa=9.137 et=5.613 tsa=10.813 va=-0.162 cs=0.104
[step 6 | 6.9s] tot=55.705 recon=4.539 sa=9.207 et=5.532 tsa=10.672 va=-0.176 cs=0.032
[step 7 | 6.7s] tot=59.428 recon=5.239 sa=9.221 et=6.308 tsa=10.750 va=-0.181 cs=0.126
[step 8 | 6.5s] tot=124.122 recon=3.637 sa=9.295 et=4.808 tsa=10.258 va=-0.180 cs=0.166
[step 9 | 6.8s] tot=59.065 recon=3.569 sa=9.533 et=4.829 tsa=10.812 va=-0.178 cs=0.159
[step 10 | 6.7s] tot=57.171 recon=4.584 sa=9.091 et=5.049 tsa=10.658 va=-0.189 cs=0.000
[step 11 | 6.3s] tot=53.703 recon=5.196 sa=9.096 et=5.832 tsa=10.738 va=-0.188 cs=0.049
[step 12 | 6.7s] tot=52.814 recon=3.690 sa=9.380 et=4.345 tsa=10.256 va=-0.183 cs=0.172
[step 13 | 6.7s] tot=50.494 recon=3.782 sa=9.439 et=4.873 tsa=10.808 va=-0.178 cs=0.129
[step 14 | 6.7s] tot=49.679 recon=4.606 sa=9.216 et=5.060 tsa=10.652 va=-0.185 cs=0.000
[step 15 | 6.6s] tot=49.815 recon=5.016 sa=9.106 et=5.609 tsa=10.725 va=-0.185 cs=0.000
[step 16 | 6.8s] tot=47.230 recon=3.760 sa=9.355 et=4.178 tsa=10.250 va=-0.180 cs=0.090
[step 17 | 6.6s] tot=48.436 recon=3.542 sa=9.370 et=4.657 tsa=10.794 va=-0.176 cs=0.033
[step 18 | 7.0s] tot=52.727 recon=4.985 sa=9.134 et=5.012 tsa=10.634 va=-0.186 cs=0.000
[step 19 | 6.8s] tot=48.519 recon=4.989 sa=9.056 et=5.591 tsa=10.692 va=-0.186 cs=0.000
[step 20 | 6.8s] tot=47.311 recon=4.105 sa=9.309 et=4.396 tsa=10.245 va=-0.181 cs=0.009
[step 21 | 6.6s] tot=46.897 recon=3.415 sa=9.265 et=4.574 tsa=10.770 va=-0.179 cs=0.000
[step 22 | 6.7s] tot=50.561 recon=5.526 sa=9.225 et=4.885 tsa=10.629 va=-0.188 cs=0.000
[step 23 | 6.9s] tot=49.184 recon=4.964 sa=9.084 et=5.415 tsa=10.674 va=-0.188 cs=0.000
[step 24 | 6.4s] tot=47.563 recon=6.038 sa=9.291 et=4.041 tsa=10.268 va=-0.181 cs=0.000
[step 25 | 6.3s] tot=46.143 recon=3.484 sa=9.089 et=4.318 tsa=10.762 va=-0.178 cs=0.000
[step 26 | 6.4s] tot=51.182 recon=7.730 sa=9.300 et=4.798 tsa=10.647 va=-0.188 cs=0.000
[step 27 | 6.4s] tot=48.543 recon=4.932 sa=9.217 et=5.339 tsa=10.677 va=-0.188 cs=0.000
[step 28 | 6.5s] tot=50.054 recon=6.523 sa=9.291 et=3.845 tsa=10.307 va=-0.183 cs=0.000
[step 29 | 6.7s] tot=47.652 recon=6.884 sa=9.002 et=4.036 tsa=10.756 va=-0.182 cs=0.000
[step 30 | 7.0s] tot=50.673 recon=7.447 sa=9.235 et=4.660 tsa=10.663 va=-0.191 cs=0.000
[step 31 | 6.8s] tot=47.293 recon=4.800 sa=9.123 et=5.195 tsa=10.676 va=-0.192 cs=0.000
[step 32 | 6.4s] tot=48.821 recon=6.581 sa=9.327 et=3.657 tsa=10.354 va=-0.187 cs=0.000
[step 33 | 6.6s] tot=47.844 recon=6.549 sa=9.125 et=3.596 tsa=10.762 va=-0.186 cs=0.000
[step 34 | 6.4s] tot=49.840 recon=7.537 sa=9.232 et=4.646 tsa=10.684 va=-0.196 cs=0.000
[step 35 | 6.4s] tot=47.458 recon=4.978 sa=9.002 et=4.822 tsa=10.681 va=-0.197 cs=0.000
[step 36 | 6.5s] tot=48.095 recon=6.583 sa=9.316 et=3.563 tsa=10.409 va=-0.191 cs=0.000
[step 37 | 6.5s] tot=46.948 recon=7.406 sa=9.063 et=3.116 tsa=10.773 va=-0.190 cs=0.000
[step 38 | 6.8s] tot=49.248 recon=7.726 sa=9.189 et=4.271 tsa=10.708 va=-0.198 cs=0.000
[step 39 | 6.5s] tot=45.898 recon=4.886 sa=9.052 et=4.372 tsa=10.695 va=-0.199 cs=0.000
[step 40 | 6.5s] tot=46.012 recon=6.571 sa=9.218 et=3.168 tsa=10.483 va=-0.192 cs=0.000
[step 41 | 6.3s] tot=47.347 recon=7.777 sa=9.141 et=2.912 tsa=10.792 va=-0.191 cs=0.000
[step 42 | 6.6s] tot=48.669 recon=7.747 sa=9.031 et=3.943 tsa=10.714 va=-0.200 cs=0.000
[step 43 | 6.5s] tot=47.022 recon=4.862 sa=9.074 et=4.247 tsa=10.704 va=-0.200 cs=0.000
[step 44 | 6.9s] tot=51.938 recon=6.689 sa=9.204 et=3.175 tsa=10.520 va=-0.193 cs=0.000
[step 45 | 6.7s] tot=46.256 recon=6.882 sa=9.156 et=2.917 tsa=10.798 va=-0.192 cs=0.000
[step 46 | 6.9s] tot=47.718 recon=7.868 sa=8.947 et=3.878 tsa=10.701 va=-0.200 cs=0.000
[step 47 | 6.7s] tot=47.864 recon=4.860 sa=9.120 et=5.451 tsa=10.705 va=-0.203 cs=0.000
[step 48 | 6.6s] tot=48.181 recon=6.511 sa=9.258 et=4.032 tsa=10.529 va=-0.197 cs=0.000
[step 49 | 6.5s] tot=47.044 recon=5.600 sa=9.113 et=4.386 tsa=10.782 va=-0.197 cs=0.000
[step 50 | 6.6s] tot=46.651 recon=7.417 sa=8.890 et=3.763 tsa=10.672 va=-0.208 cs=0.000
[step 51 | 6.4s] tot=46.120 recon=4.791 sa=9.140 et=4.544 tsa=10.705 va=-0.210 cs=0.000
[step 52 | 6.4s] tot=45.525 recon=5.872 sa=9.240 et=3.308 tsa=10.535 va=-0.205 cs=0.000
[step 53 | 6.7s] tot=46.228 recon=6.025 sa=9.056 et=3.544 tsa=10.764 va=-0.205 cs=0.000
[step 54 | 6.5s] tot=45.687 recon=7.075 sa=8.867 et=3.546 tsa=10.648 va=-0.216 cs=0.000
[step 55 | 6.2s] tot=44.940 recon=4.752 sa=9.072 et=4.160 tsa=10.689 va=-0.218 cs=0.000
[step 56 | 6.2s] tot=43.719 recon=4.915 sa=9.097 et=2.926 tsa=10.503 va=-0.213 cs=0.000
[step 57 | 6.4s] tot=43.905 recon=5.389 sa=9.061 et=2.558 tsa=10.751 va=-0.213 cs=0.000
[step 58 | 6.7s] tot=43.939 recon=5.272 sa=8.868 et=3.474 tsa=10.631 va=-0.223 cs=0.000
[step 59 | 6.6s] tot=44.196 recon=4.783 sa=9.020 et=3.839 tsa=10.682 va=-0.225 cs=0.000

[done] total train time: 398.5s avg/step=6.6s
[done] checkpoint saved: ckpt/v344_trained.pt (196 tensors)
Expand Down
Loading