Training in progress, step 400, checkpoint

last-checkpoint/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "fxmarty/tiny-dummy-qwen2",
+  "_name_or_path": "peft-internal-testing/tiny-dummy-qwen2",
   "architectures": [
     "Qwen2ForCausalLM"
   ],

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89fe48d90293c6de8a412c54d1431a33bdd6255c0ce7cf015902d4131b92c913
+oid sha256:d74a2d30e9b5eaccdc8860d90690f3f6102f6de30d1c8b6b8ec10411ff53d836
 size 2433024

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf7ecc2f0c8d284946a757ca8f3ded322c622c5f84f3780795fb07bff724c053
+oid sha256:2b7c01f6981a3ced8f46ae6b6f209864546e601aaf3a635f661d708f4b18efaf
 size 2498406

last-checkpoint/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9feae33b2fec0a6229240e7adaee6ecc8f5cfdf1a8bd0e827b1d8a241424e3c0
+size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f0c7ff54b85ffef1dcbffa5d8d256d7b7e02cf6f2a611b338e53d605c1ee098
+oid sha256:a673aaf85c0fe6b6c29cb8f3e7dbd829eef637110e4ad9a775f3fcf001c92591
 size 1064

last-checkpoint/trainer_state.json

@@ -1,203 +1,323 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.1893491124260355,
-  "eval_steps": 20,
-  "global_step": 160,
+  "epoch": 0.11837821840781296,
+  "eval_steps": 200,
+  "global_step": 400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.001183431952662722,
-      "eval_loss": 11.929322242736816,
-      "eval_runtime": 3.5424,
-      "eval_samples_per_second": 424.005,
-      "eval_steps_per_second": 26.536,
+      "epoch": 0.0002959455460195324,
+      "eval_loss": 11.929207801818848,
+      "eval_runtime": 36.681,
+      "eval_samples_per_second": 40.948,
+      "eval_steps_per_second": 10.251,
       "step": 1
     },
     {
-      "epoch": 0.011834319526627219,
-      "grad_norm": 0.19140625,
+      "epoch": 0.002959455460195324,
+      "grad_norm": 0.212890625,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 11.9299,
+      "loss": 11.9295,
       "step": 10
     },
     {
-      "epoch": 0.023668639053254437,
-      "grad_norm": 0.12158203125,
+      "epoch": 0.005918910920390648,
+      "grad_norm": 0.263671875,
       "learning_rate": 3.2000000000000005e-05,
-      "loss": 11.9289,
+      "loss": 11.9301,
       "step": 20
     },
     {
-      "epoch": 0.023668639053254437,
-      "eval_loss": 11.92916488647461,
-      "eval_runtime": 3.5376,
-      "eval_samples_per_second": 424.587,
-      "eval_steps_per_second": 26.572,
-      "step": 20
-    },
-    {
-      "epoch": 0.03550295857988166,
-      "grad_norm": 0.251953125,
+      "epoch": 0.008878366380585973,
+      "grad_norm": 0.375,
       "learning_rate": 4.8e-05,
-      "loss": 11.9288,
+      "loss": 11.9292,
       "step": 30
     },
     {
-      "epoch": 0.047337278106508875,
-      "grad_norm": 0.1435546875,
+      "epoch": 0.011837821840781295,
+      "grad_norm": 0.47265625,
       "learning_rate": 6.400000000000001e-05,
-      "loss": 11.9293,
+      "loss": 11.926,
       "step": 40
     },
     {
-      "epoch": 0.047337278106508875,
-      "eval_loss": 11.928914070129395,
-      "eval_runtime": 3.5225,
-      "eval_samples_per_second": 426.404,
-      "eval_steps_per_second": 26.686,
-      "step": 40
-    },
-    {
-      "epoch": 0.05917159763313609,
-      "grad_norm": 0.373046875,
+      "epoch": 0.01479727730097662,
+      "grad_norm": 0.84375,
       "learning_rate": 8e-05,
-      "loss": 11.9265,
+      "loss": 11.9345,
       "step": 50
     },
     {
-      "epoch": 0.07100591715976332,
-      "grad_norm": 0.158203125,
+      "epoch": 0.017756732761171946,
+      "grad_norm": 0.244140625,
       "learning_rate": 9.6e-05,
-      "loss": 11.9267,
-      "step": 60
-    },
-    {
-      "epoch": 0.07100591715976332,
-      "eval_loss": 11.92799186706543,
-      "eval_runtime": 3.5615,
-      "eval_samples_per_second": 421.727,
-      "eval_steps_per_second": 26.393,
+      "loss": 11.9293,
       "step": 60
     },
     {
-      "epoch": 0.08284023668639054,
-      "grad_norm": 0.146484375,
+      "epoch": 0.020716188221367268,
+      "grad_norm": 0.255859375,
       "learning_rate": 0.00011200000000000001,
-      "loss": 11.9275,
+      "loss": 11.9289,
       "step": 70
     },
     {
-      "epoch": 0.09467455621301775,
-      "grad_norm": 0.2333984375,
+      "epoch": 0.02367564368156259,
+      "grad_norm": 0.34375,
       "learning_rate": 0.00012800000000000002,
-      "loss": 11.9256,
+      "loss": 11.9291,
       "step": 80
     },
     {
-      "epoch": 0.09467455621301775,
-      "eval_loss": 11.925419807434082,
-      "eval_runtime": 3.5961,
-      "eval_samples_per_second": 417.672,
-      "eval_steps_per_second": 26.139,
-      "step": 80
-    },
-    {
-      "epoch": 0.10650887573964497,
-      "grad_norm": 0.236328125,
+      "epoch": 0.026635099141757917,
+      "grad_norm": 0.453125,
       "learning_rate": 0.000144,
-      "loss": 11.9219,
+      "loss": 11.9271,
       "step": 90
     },
     {
-      "epoch": 0.11834319526627218,
-      "grad_norm": 0.51953125,
+      "epoch": 0.02959455460195324,
+      "grad_norm": 0.80859375,
       "learning_rate": 0.00016,
-      "loss": 11.9151,
-      "step": 100
-    },
-    {
-      "epoch": 0.11834319526627218,
-      "eval_loss": 11.913871765136719,
-      "eval_runtime": 3.5719,
-      "eval_samples_per_second": 420.504,
-      "eval_steps_per_second": 26.316,
+      "loss": 11.9272,
       "step": 100
     },
     {
-      "epoch": 0.1301775147928994,
-      "grad_norm": 0.494140625,
+      "epoch": 0.032554010062148565,
+      "grad_norm": 0.23046875,
       "learning_rate": 0.00017600000000000002,
-      "loss": 11.9056,
+      "loss": 11.9286,
       "step": 110
     },
     {
-      "epoch": 0.14201183431952663,
-      "grad_norm": 0.6171875,
+      "epoch": 0.03551346552234389,
+      "grad_norm": 0.322265625,
       "learning_rate": 0.000192,
-      "loss": 11.8786,
+      "loss": 11.9239,
       "step": 120
     },
     {
-      "epoch": 0.14201183431952663,
-      "eval_loss": 11.85535717010498,
-      "eval_runtime": 3.5396,
-      "eval_samples_per_second": 424.345,
-      "eval_steps_per_second": 26.557,
-      "step": 120
-    },
-    {
-      "epoch": 0.15384615384615385,
-      "grad_norm": 0.59765625,
+      "epoch": 0.03847292098253921,
+      "grad_norm": 0.4296875,
       "learning_rate": 0.0001999978128380225,
-      "loss": 11.8261,
+      "loss": 11.929,
       "step": 130
     },
     {
-      "epoch": 0.16568047337278108,
-      "grad_norm": 0.4921875,
+      "epoch": 0.041432376442734536,
+      "grad_norm": 0.5,
       "learning_rate": 0.0001999803161162393,
-      "loss": 11.7729,
-      "step": 140
-    },
-    {
-      "epoch": 0.16568047337278108,
-      "eval_loss": 11.73879337310791,
-      "eval_runtime": 3.4959,
-      "eval_samples_per_second": 429.652,
-      "eval_steps_per_second": 26.889,
+      "loss": 11.9268,
       "step": 140
     },
     {
-      "epoch": 0.17751479289940827,
-      "grad_norm": 0.6171875,
+      "epoch": 0.04439183190292986,
+      "grad_norm": 1.2578125,
       "learning_rate": 0.00019994532573409262,
-      "loss": 11.714,
+      "loss": 11.9198,
       "step": 150
     },
     {
-      "epoch": 0.1893491124260355,
-      "grad_norm": 0.390625,
+      "epoch": 0.04735128736312518,
+      "grad_norm": 0.400390625,
       "learning_rate": 0.00019989284781388617,
-      "loss": 11.6588,
+      "loss": 11.9145,
       "step": 160
     },
     {
-      "epoch": 0.1893491124260355,
-      "eval_loss": 11.629658699035645,
-      "eval_runtime": 3.5771,
-      "eval_samples_per_second": 419.895,
-      "eval_steps_per_second": 26.278,
-      "step": 160
+      "epoch": 0.05031074282332051,
+      "grad_norm": 0.5234375,
+      "learning_rate": 0.00019982289153773646,
+      "loss": 11.9004,
+      "step": 170
+    },
+    {
+      "epoch": 0.053270198283515834,
+      "grad_norm": 0.796875,
+      "learning_rate": 0.00019973546914596623,
+      "loss": 11.8792,
+      "step": 180
+    },
+    {
+      "epoch": 0.05622965374371116,
+      "grad_norm": 0.7109375,
+      "learning_rate": 0.00019963059593496268,
+      "loss": 11.8308,
+      "step": 190
+    },
+    {
+      "epoch": 0.05918910920390648,
+      "grad_norm": 0.7734375,
+      "learning_rate": 0.00019950829025450114,
+      "loss": 11.78,
+      "step": 200
+    },
+    {
+      "epoch": 0.05918910920390648,
+      "eval_loss": 11.750052452087402,
+      "eval_runtime": 37.1832,
+      "eval_samples_per_second": 40.395,
+      "eval_steps_per_second": 10.112,
+      "step": 200
+    },
+    {
+      "epoch": 0.062148564664101805,
+      "grad_norm": 0.451171875,
+      "learning_rate": 0.0001993685735045343,
+      "loss": 11.7099,
+      "step": 210
+    },
+    {
+      "epoch": 0.06510802012429713,
+      "grad_norm": 0.435546875,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 11.6724,
+      "step": 220
+    },
+    {
+      "epoch": 0.06806747558449246,
+      "grad_norm": 0.33203125,
+      "learning_rate": 0.000199037007623783,
+      "loss": 11.6468,
+      "step": 230
+    },
+    {
+      "epoch": 0.07102693104468778,
+      "grad_norm": 0.625,
+      "learning_rate": 0.00019884521650742715,
+      "loss": 11.6114,
+      "step": 240
+    },
+    {
+      "epoch": 0.0739863865048831,
+      "grad_norm": 0.9296875,
+      "learning_rate": 0.00019863613034027224,
+      "loss": 11.6611,
+      "step": 250
+    },
+    {
+      "epoch": 0.07694584196507842,
+      "grad_norm": 0.482421875,
+      "learning_rate": 0.0001984097857063434,
+      "loss": 11.4853,
+      "step": 260
+    },
+    {
+      "epoch": 0.07990529742527375,
+      "grad_norm": 0.41796875,
+      "learning_rate": 0.0001981662222093976,
+      "loss": 11.4982,
+      "step": 270
+    },
+    {
+      "epoch": 0.08286475288546907,
+      "grad_norm": 0.59375,
+      "learning_rate": 0.00019790548246599447,
+      "loss": 11.4479,
+      "step": 280
+    },
+    {
+      "epoch": 0.0858242083456644,
+      "grad_norm": 1.0078125,
+      "learning_rate": 0.00019762761209803927,
+      "loss": 11.3621,
+      "step": 290
+    },
+    {
+      "epoch": 0.08878366380585972,
+      "grad_norm": 1.921875,
+      "learning_rate": 0.0001973326597248006,
+      "loss": 11.3921,
+      "step": 300
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.400390625,
+      "learning_rate": 0.00019702067695440332,
+      "loss": 11.2354,
+      "step": 310
+    },
+    {
+      "epoch": 0.09470257472625036,
+      "grad_norm": 0.490234375,
+      "learning_rate": 0.00019669171837479873,
+      "loss": 11.2262,
+      "step": 320
+    },
+    {
+      "epoch": 0.09766203018644569,
+      "grad_norm": 0.5234375,
+      "learning_rate": 0.00019634584154421317,
+      "loss": 11.1823,
+      "step": 330
+    },
+    {
+      "epoch": 0.10062148564664102,
+      "grad_norm": 0.46875,
+      "learning_rate": 0.00019598310698107702,
+      "loss": 11.1421,
+      "step": 340
+    },
+    {
+      "epoch": 0.10358094110683634,
+      "grad_norm": 1.203125,
+      "learning_rate": 0.00019560357815343577,
+      "loss": 11.1708,
+      "step": 350
+    },
+    {
+      "epoch": 0.10654039656703167,
+      "grad_norm": 0.52734375,
+      "learning_rate": 0.00019520732146784491,
+      "loss": 11.1091,
+      "step": 360
+    },
+    {
+      "epoch": 0.109499852027227,
+      "grad_norm": 0.67578125,
+      "learning_rate": 0.0001947944062577507,
+      "loss": 11.0413,
+      "step": 370
+    },
+    {
+      "epoch": 0.11245930748742232,
+      "grad_norm": 0.44921875,
+      "learning_rate": 0.00019436490477135878,
+      "loss": 11.0481,
+      "step": 380
+    },
+    {
+      "epoch": 0.11541876294761765,
+      "grad_norm": 0.640625,
+      "learning_rate": 0.00019391889215899299,
+      "loss": 11.1447,
+      "step": 390
+    },
+    {
+      "epoch": 0.11837821840781296,
+      "grad_norm": 1.203125,
+      "learning_rate": 0.0001934564464599461,
+      "loss": 10.974,
+      "step": 400
+    },
+    {
+      "epoch": 0.11837821840781296,
+      "eval_loss": 11.000988006591797,
+      "eval_runtime": 37.2684,
+      "eval_samples_per_second": 40.302,
+      "eval_steps_per_second": 10.089,
+      "step": 400
     }
   ],
   "logging_steps": 10,
   "max_steps": 2500,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
-  "save_steps": 40,
+  "num_train_epochs": 1,
+  "save_steps": 400,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -210,8 +330,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 62711857152.0,
-  "train_batch_size": 2,
+  "total_flos": 39262224384.0,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20af897fff432dad9da1232d2f5e6ab4040448b601443dcf983c39a3e47a3b26
-size 6648
+oid sha256:6cafd6770e4d3416c27e43a8308e52009a8c0c1fff65ee59e17f0872cd1b39e9
+size 6840