rakhman-llm commited on
Commit
f645c58
·
verified ·
1 Parent(s): df8b86d

Training in progress, step 16000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:47ec1fd68430fdacb80810ecd09adfa63b256edd752390b582d7134186c11fe5
3
  size 891558696
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39065861c3b099700bb000236a9e498a039343f48398b98c2fb9a14bb096b6c1
3
  size 891558696
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7dedc049b8b109e2a1e1d8ca3aa0a4db96cc41f55ad87c5be1b4f1d3b32922bb
3
  size 1783272762
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e12bc163cfe02cd564c638e298b40f4155a4f87b8f01adeb81746cbd797e5ef
3
  size 1783272762
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dc7b29b924630649498d282773659d36f4c79fef1453c3983d39d7e7e53066c4
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eebbe4891f802a9c170f9968f3ddf4d71a0eff2d959ee8775998aa9a7034155
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0adcdf1185fa7d1b89a354568f52beda1398968ea15296ac3bc948fc9b42fcfd
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6949e36257fd6db2d60885342ebdb2dc7d96352bf8e47bc5271716a48f4cbc6e
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": 0.08158940076828003,
3
  "best_model_checkpoint": "./fine-tuned/checkpoint-12500",
4
- "epoch": 2.48,
5
  "eval_steps": 500,
6
- "global_step": 15500,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -2425,6 +2425,84 @@
2425
  "eval_samples_per_second": 17.134,
2426
  "eval_steps_per_second": 2.142,
2427
  "step": 15500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2428
  }
2429
  ],
2430
  "logging_steps": 50,
@@ -2444,7 +2522,7 @@
2444
  "attributes": {}
2445
  }
2446
  },
2447
- "total_flos": 7.551077842944e+16,
2448
  "train_batch_size": 8,
2449
  "trial_name": null,
2450
  "trial_params": null
 
1
  {
2
  "best_metric": 0.08158940076828003,
3
  "best_model_checkpoint": "./fine-tuned/checkpoint-12500",
4
+ "epoch": 2.56,
5
  "eval_steps": 500,
6
+ "global_step": 16000,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
2425
  "eval_samples_per_second": 17.134,
2426
  "eval_steps_per_second": 2.142,
2427
  "step": 15500
2428
+ },
2429
+ {
2430
+ "epoch": 2.488,
2431
+ "grad_norm": 6472.25537109375,
2432
+ "learning_rate": 5.12e-06,
2433
+ "loss": 0.0496,
2434
+ "step": 15550
2435
+ },
2436
+ {
2437
+ "epoch": 2.496,
2438
+ "grad_norm": 6369.4833984375,
2439
+ "learning_rate": 5.04e-06,
2440
+ "loss": 0.0518,
2441
+ "step": 15600
2442
+ },
2443
+ {
2444
+ "epoch": 2.504,
2445
+ "grad_norm": 8784.1083984375,
2446
+ "learning_rate": 4.96e-06,
2447
+ "loss": 0.0547,
2448
+ "step": 15650
2449
+ },
2450
+ {
2451
+ "epoch": 2.512,
2452
+ "grad_norm": 8509.6650390625,
2453
+ "learning_rate": 4.88e-06,
2454
+ "loss": 0.0555,
2455
+ "step": 15700
2456
+ },
2457
+ {
2458
+ "epoch": 2.52,
2459
+ "grad_norm": 7856.84716796875,
2460
+ "learning_rate": 4.800000000000001e-06,
2461
+ "loss": 0.0513,
2462
+ "step": 15750
2463
+ },
2464
+ {
2465
+ "epoch": 2.528,
2466
+ "grad_norm": 6816.51123046875,
2467
+ "learning_rate": 4.72e-06,
2468
+ "loss": 0.0493,
2469
+ "step": 15800
2470
+ },
2471
+ {
2472
+ "epoch": 2.536,
2473
+ "grad_norm": 6773.06884765625,
2474
+ "learning_rate": 4.6400000000000005e-06,
2475
+ "loss": 0.05,
2476
+ "step": 15850
2477
+ },
2478
+ {
2479
+ "epoch": 2.544,
2480
+ "grad_norm": 9726.3818359375,
2481
+ "learning_rate": 4.56e-06,
2482
+ "loss": 0.0518,
2483
+ "step": 15900
2484
+ },
2485
+ {
2486
+ "epoch": 2.552,
2487
+ "grad_norm": 8707.6591796875,
2488
+ "learning_rate": 4.48e-06,
2489
+ "loss": 0.0499,
2490
+ "step": 15950
2491
+ },
2492
+ {
2493
+ "epoch": 2.56,
2494
+ "grad_norm": 4772.958984375,
2495
+ "learning_rate": 4.4e-06,
2496
+ "loss": 0.0522,
2497
+ "step": 16000
2498
+ },
2499
+ {
2500
+ "epoch": 2.56,
2501
+ "eval_loss": 0.08175843954086304,
2502
+ "eval_runtime": 116.8011,
2503
+ "eval_samples_per_second": 17.123,
2504
+ "eval_steps_per_second": 2.14,
2505
+ "step": 16000
2506
  }
2507
  ],
2508
  "logging_steps": 50,
 
2522
  "attributes": {}
2523
  }
2524
  },
2525
+ "total_flos": 7.794660999168e+16,
2526
  "train_batch_size": 8,
2527
  "trial_name": null,
2528
  "trial_params": null