qingy2024 commited on
Commit
ea37448
·
verified ·
1 Parent(s): 7c959b8

Upload checkpoint 170

Browse files
Files changed (7) hide show
  1. README.md +4 -4
  2. adapter_model.safetensors +1 -1
  3. loss.png +2 -2
  4. optimizer.pt +1 -1
  5. rng_state.pth +1 -1
  6. scheduler.pt +1 -1
  7. trainer_state.json +152 -2
README.md CHANGED
@@ -2,7 +2,7 @@
2
  base_model: InfiniAILab/OpenR1-Qwen-3B-SFT-Instruct
3
  library_name: peft
4
  ---
5
- # T2 3B Instruct (Step 160 Checkpoint)
6
 
7
  > [!NOTE]
8
  > Training in progress...
@@ -15,11 +15,11 @@ library_name: peft
15
  </head>
16
  <body>
17
  <div style="width: 100%; background-color: #e0e0e0; border-radius: 25px; overflow: hidden; margin: 20px 0;">
18
- <div style="height: 30px; width: 0.62%; background-color: #44965a; text-align: center; line-height: 30px; color: white; border-radius: 25px 0 0 25px;">
19
- <!-- 0.6% -->
20
  </div>
21
  </div>
22
- <p style="font-family: Arial, sans-serif; font-size: 16px;">Progress: 160 out of 25777 steps</p>
23
  </body>
24
  </html>
25
 
 
2
  base_model: InfiniAILab/OpenR1-Qwen-3B-SFT-Instruct
3
  library_name: peft
4
  ---
5
+ # T2 3B Instruct (Step 170 Checkpoint)
6
 
7
  > [!NOTE]
8
  > Training in progress...
 
15
  </head>
16
  <body>
17
  <div style="width: 100%; background-color: #e0e0e0; border-radius: 25px; overflow: hidden; margin: 20px 0;">
18
+ <div style="height: 30px; width: 0.66%; background-color: #44965a; text-align: center; line-height: 30px; color: white; border-radius: 25px 0 0 25px;">
19
+ <!-- 0.7% -->
20
  </div>
21
  </div>
22
+ <p style="font-family: Arial, sans-serif; font-size: 16px;">Progress: 170 out of 25777 steps</p>
23
  </body>
24
  </html>
25
 
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d4b893a27f3d27ba676c438a5f86ddc4d9336144ad21d9942852f78c774cbd37
3
  size 479005064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f8a596570de87e145529afffd343178c86321d4ca8807e22140f8266ecdc299
3
  size 479005064
loss.png CHANGED

Git LFS Details

  • SHA256: c164ea0896f681bf5633f59777f48c5570d56c9e2cd3ea0a2e23de1294d3930f
  • Pointer size: 131 Bytes
  • Size of remote file: 209 kB

Git LFS Details

  • SHA256: 941ec638e7e9e89a5ebfecf497f39ac2f4dd342faa2c504ab84c0ae7460c6578
  • Pointer size: 131 Bytes
  • Size of remote file: 211 kB
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a7e1aed056922b7461c03caf8d4edb1d08a490938aa6297545a6879163f88f04
3
  size 245114274
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c55b7e32f9b1503124db65c231342036e4beb0774dfba0c0ac92df16299fbdf1
3
  size 245114274
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4974c6adf2bc23560a44eef6c2e199ff72657aa6de3a59402a1de0a6d022f85d
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d82e12ca8e174d0b8a92b1ba42160325264a4e6ae98b3bfda7e1a54d3203a344
3
  size 14244
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:47e8247f95d57e32f07911dcc0ac85cb8cecf4ded8c927587c764dc8757c1e8a
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2280924964fead3114ca3d9e6ddaed8cdc190e64f2a1a0a3e012050d02f0af8
3
  size 1064
trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.0062069634371060035,
6
  "eval_steps": 500,
7
- "global_step": 160,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -2408,6 +2408,156 @@
2408
  "rewards/strict_format_reward_func": 0.03750000009313226,
2409
  "rewards/thinkcount_reward_func": 0.031250000931322575,
2410
  "step": 160
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2411
  }
2412
  ],
2413
  "logging_steps": 1,
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.0065948986519251285,
6
  "eval_steps": 500,
7
+ "global_step": 170,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
2408
  "rewards/strict_format_reward_func": 0.03750000009313226,
2409
  "rewards/thinkcount_reward_func": 0.031250000931322575,
2410
  "step": 160
2411
+ },
2412
+ {
2413
+ "completion_length": 8114.5,
2414
+ "epoch": 0.0062457569585879155,
2415
+ "grad_norm": 5.941150665283203,
2416
+ "kl": 0.0077293055364862084,
2417
+ "learning_rate": 3.1031807602792863e-07,
2418
+ "loss": 0.0003,
2419
+ "reward": 0.5584961185231805,
2420
+ "reward_std": 0.2153224777430296,
2421
+ "rewards/correctness_reward_func": 0.48193359375,
2422
+ "rewards/soft_format_reward_func": 0.018750000279396772,
2423
+ "rewards/strict_format_reward_func": 0.015625000465661287,
2424
+ "rewards/thinkcount_reward_func": 0.042187499813735485,
2425
+ "step": 161
2426
+ },
2427
+ {
2428
+ "completion_length": 8103.0,
2429
+ "epoch": 0.006284550480069828,
2430
+ "grad_norm": 5.055596828460693,
2431
+ "kl": 0.011864460189826787,
2432
+ "learning_rate": 3.122575640031032e-07,
2433
+ "loss": 0.0005,
2434
+ "reward": 0.8435171665623784,
2435
+ "reward_std": 0.7845239788293839,
2436
+ "rewards/correctness_reward_func": 0.7392578125,
2437
+ "rewards/soft_format_reward_func": 0.03281250037252903,
2438
+ "rewards/strict_format_reward_func": 0.037500000558793545,
2439
+ "rewards/thinkcount_reward_func": 0.033946875017136335,
2440
+ "step": 162
2441
+ },
2442
+ {
2443
+ "completion_length": 8078.0,
2444
+ "epoch": 0.006323344001551741,
2445
+ "grad_norm": 4.159585475921631,
2446
+ "kl": 0.00777029967866838,
2447
+ "learning_rate": 3.141970519782778e-07,
2448
+ "loss": 0.0003,
2449
+ "reward": 1.059982754290104,
2450
+ "reward_std": 0.6603782661259174,
2451
+ "rewards/correctness_reward_func": 0.9677734375,
2452
+ "rewards/soft_format_reward_func": 0.025000000023283064,
2453
+ "rewards/strict_format_reward_func": 0.03437500004656613,
2454
+ "rewards/thinkcount_reward_func": 0.03283437585923821,
2455
+ "step": 163
2456
+ },
2457
+ {
2458
+ "completion_length": 8094.0,
2459
+ "epoch": 0.006362137523033653,
2460
+ "grad_norm": 3.46872615814209,
2461
+ "kl": 0.008031659293919802,
2462
+ "learning_rate": 3.1613653995345233e-07,
2463
+ "loss": 0.0003,
2464
+ "reward": 0.4744140300899744,
2465
+ "reward_std": 0.5823787748813629,
2466
+ "rewards/correctness_reward_func": 0.4072265625,
2467
+ "rewards/soft_format_reward_func": 0.010937500279396772,
2468
+ "rewards/strict_format_reward_func": 0.012500000186264515,
2469
+ "rewards/thinkcount_reward_func": 0.04375000111758709,
2470
+ "step": 164
2471
+ },
2472
+ {
2473
+ "completion_length": 7937.46875,
2474
+ "epoch": 0.006400931044515566,
2475
+ "grad_norm": 8.129719734191895,
2476
+ "kl": 0.008240840281359851,
2477
+ "learning_rate": 3.180760279286269e-07,
2478
+ "loss": 0.0003,
2479
+ "reward": 1.3455124124884605,
2480
+ "reward_std": 0.754421828314662,
2481
+ "rewards/correctness_reward_func": 1.220703125,
2482
+ "rewards/soft_format_reward_func": 0.043750000186264515,
2483
+ "rewards/strict_format_reward_func": 0.05937500111758709,
2484
+ "rewards/thinkcount_reward_func": 0.021684375125914812,
2485
+ "step": 165
2486
+ },
2487
+ {
2488
+ "completion_length": 8124.0,
2489
+ "epoch": 0.006439724565997478,
2490
+ "grad_norm": 3.3704938888549805,
2491
+ "kl": 0.007638707873411477,
2492
+ "learning_rate": 3.2001551590380137e-07,
2493
+ "loss": 0.0003,
2494
+ "reward": 0.3531265389174223,
2495
+ "reward_std": 0.5546747129410505,
2496
+ "rewards/correctness_reward_func": 0.2763671875,
2497
+ "rewards/soft_format_reward_func": 0.014062500442378223,
2498
+ "rewards/strict_format_reward_func": 0.021875000558793545,
2499
+ "rewards/thinkcount_reward_func": 0.04082187544554472,
2500
+ "step": 166
2501
+ },
2502
+ {
2503
+ "completion_length": 8111.25,
2504
+ "epoch": 0.006478518087479391,
2505
+ "grad_norm": 5.47683048248291,
2506
+ "kl": 0.0084478979697451,
2507
+ "learning_rate": 3.21955003878976e-07,
2508
+ "loss": 0.0003,
2509
+ "reward": 0.6906819827854633,
2510
+ "reward_std": 0.7905658520758152,
2511
+ "rewards/correctness_reward_func": 0.60400390625,
2512
+ "rewards/soft_format_reward_func": 0.01875000016298145,
2513
+ "rewards/strict_format_reward_func": 0.034375000512227416,
2514
+ "rewards/thinkcount_reward_func": 0.03355312626808882,
2515
+ "step": 167
2516
+ },
2517
+ {
2518
+ "completion_length": 8062.25,
2519
+ "epoch": 0.006517311608961304,
2520
+ "grad_norm": 5.1698899269104,
2521
+ "kl": 0.009413379244506359,
2522
+ "learning_rate": 3.238944918541505e-07,
2523
+ "loss": 0.0004,
2524
+ "reward": 0.2749023260548711,
2525
+ "reward_std": 0.47187073063105345,
2526
+ "rewards/correctness_reward_func": 0.20458984375,
2527
+ "rewards/soft_format_reward_func": 0.01406250020954758,
2528
+ "rewards/strict_format_reward_func": 0.012500000186264515,
2529
+ "rewards/thinkcount_reward_func": 0.043750000186264515,
2530
+ "step": 168
2531
+ },
2532
+ {
2533
+ "completion_length": 8110.25,
2534
+ "epoch": 0.006556105130443216,
2535
+ "grad_norm": 4.802003383636475,
2536
+ "kl": 0.007399239810183644,
2537
+ "learning_rate": 3.2583397982932507e-07,
2538
+ "loss": 0.0003,
2539
+ "reward": 0.7043304797261953,
2540
+ "reward_std": 0.8031427129171789,
2541
+ "rewards/correctness_reward_func": 0.61279296875,
2542
+ "rewards/soft_format_reward_func": 0.02500000048894435,
2543
+ "rewards/strict_format_reward_func": 0.031250000931322575,
2544
+ "rewards/thinkcount_reward_func": 0.035287500359117985,
2545
+ "step": 169
2546
+ },
2547
+ {
2548
+ "completion_length": 8109.25,
2549
+ "epoch": 0.0065948986519251285,
2550
+ "grad_norm": 3.9394149780273438,
2551
+ "kl": 0.008788365870714188,
2552
+ "learning_rate": 3.277734678044996e-07,
2553
+ "loss": 0.0004,
2554
+ "reward": 0.6958187464624643,
2555
+ "reward_std": 0.5882456600666046,
2556
+ "rewards/correctness_reward_func": 0.61328125,
2557
+ "rewards/soft_format_reward_func": 0.018749999813735485,
2558
+ "rewards/strict_format_reward_func": 0.02500000037252903,
2559
+ "rewards/thinkcount_reward_func": 0.03878750093281269,
2560
+ "step": 170
2561
  }
2562
  ],
2563
  "logging_steps": 1,