Upload folder using huggingface_hub

flyte_training_config.json

@@ -1 +1 @@
-{"base_model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 1, "max_steps": 500, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 16, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}
+{"base_model": "togethercomputer/RedPajama-INCITE-Base-3B-v1", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 1, "max_steps": 100, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 16, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f6e921b4581ba057c3b200ae0688f99915374b4184b5dc7876c4450b8eee8cf
+oid sha256:e63cb70257a1b71bfddd7ac332cf76d74e1d295fe8c1bd2752e1b18938257b47
 size 5686106713

trainer_state.json

@@ -1,325 +1,85 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 333.3333333333333,
-  "global_step": 500,
+  "epoch": 66.66666666666667,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 6.67,
-      "learning_rate": 1.5357481488588927e-05,
-      "loss": 1.9835,
+      "learning_rate": 2e-05,
+      "loss": 1.9669,
       "step": 10
     },
     {
       "epoch": 13.33,
       "learning_rate": 2e-05,
-      "loss": 1.0763,
+      "loss": 0.9447,
       "step": 20
     },
     {
       "epoch": 20.0,
       "learning_rate": 2e-05,
-      "loss": 0.2284,
+      "loss": 0.1908,
       "step": 30
     },
     {
       "epoch": 26.67,
       "learning_rate": 2e-05,
-      "loss": 0.0691,
+      "loss": 0.0666,
       "step": 40
     },
     {
       "epoch": 33.33,
       "learning_rate": 2e-05,
-      "loss": 0.0442,
+      "loss": 0.0441,
       "step": 50
     },
     {
       "epoch": 40.0,
       "learning_rate": 2e-05,
-      "loss": 0.0319,
+      "loss": 0.0329,
       "step": 60
     },
     {
       "epoch": 46.67,
       "learning_rate": 2e-05,
-      "loss": 0.0243,
+      "loss": 0.0251,
       "step": 70
     },
     {
       "epoch": 53.33,
       "learning_rate": 2e-05,
-      "loss": 0.0186,
+      "loss": 0.019,
       "step": 80
     },
     {
       "epoch": 60.0,
       "learning_rate": 2e-05,
-      "loss": 0.0149,
+      "loss": 0.0166,
       "step": 90
     },
     {
       "epoch": 66.67,
       "learning_rate": 2e-05,
-      "loss": 0.0119,
+      "loss": 0.0133,
       "step": 100
     },
     {
-      "epoch": 73.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0103,
-      "step": 110
-    },
-    {
-      "epoch": 80.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0091,
-      "step": 120
-    },
-    {
-      "epoch": 86.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0081,
-      "step": 130
-    },
-    {
-      "epoch": 93.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0073,
-      "step": 140
-    },
-    {
-      "epoch": 100.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0064,
-      "step": 150
-    },
-    {
-      "epoch": 106.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0064,
-      "step": 160
-    },
-    {
-      "epoch": 113.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0059,
-      "step": 170
-    },
-    {
-      "epoch": 120.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0057,
-      "step": 180
-    },
-    {
-      "epoch": 126.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0054,
-      "step": 190
-    },
-    {
-      "epoch": 133.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0052,
-      "step": 200
-    },
-    {
-      "epoch": 140.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0049,
-      "step": 210
-    },
-    {
-      "epoch": 146.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0046,
-      "step": 220
-    },
-    {
-      "epoch": 153.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0045,
-      "step": 230
-    },
-    {
-      "epoch": 160.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0044,
-      "step": 240
-    },
-    {
-      "epoch": 166.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0043,
-      "step": 250
-    },
-    {
-      "epoch": 173.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0044,
-      "step": 260
-    },
-    {
-      "epoch": 180.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0043,
-      "step": 270
-    },
-    {
-      "epoch": 186.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0044,
-      "step": 280
-    },
-    {
-      "epoch": 193.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0041,
-      "step": 290
-    },
-    {
-      "epoch": 200.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0042,
-      "step": 300
-    },
-    {
-      "epoch": 206.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0043,
-      "step": 310
-    },
-    {
-      "epoch": 213.33,
-      "learning_rate": 2e-05,
-      "loss": 0.004,
-      "step": 320
-    },
-    {
-      "epoch": 220.0,
-      "learning_rate": 2e-05,
-      "loss": 0.004,
-      "step": 330
-    },
-    {
-      "epoch": 226.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0041,
-      "step": 340
-    },
-    {
-      "epoch": 233.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0039,
-      "step": 350
-    },
-    {
-      "epoch": 240.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0039,
-      "step": 360
-    },
-    {
-      "epoch": 246.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 370
-    },
-    {
-      "epoch": 253.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0039,
-      "step": 380
-    },
-    {
-      "epoch": 260.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0039,
-      "step": 390
-    },
-    {
-      "epoch": 266.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 400
-    },
-    {
-      "epoch": 273.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 410
-    },
-    {
-      "epoch": 280.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 420
-    },
-    {
-      "epoch": 286.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0037,
-      "step": 430
-    },
-    {
-      "epoch": 293.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 440
-    },
-    {
-      "epoch": 300.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0039,
-      "step": 450
-    },
-    {
-      "epoch": 306.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0037,
-      "step": 460
-    },
-    {
-      "epoch": 313.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0038,
-      "step": 470
-    },
-    {
-      "epoch": 320.0,
-      "learning_rate": 2e-05,
-      "loss": 0.0036,
-      "step": 480
-    },
-    {
-      "epoch": 326.67,
-      "learning_rate": 2e-05,
-      "loss": 0.0037,
-      "step": 490
-    },
-    {
-      "epoch": 333.33,
-      "learning_rate": 2e-05,
-      "loss": 0.0037,
-      "step": 500
-    },
-    {
-      "epoch": 333.33,
-      "step": 500,
-      "total_flos": 210359990353920.0,
-      "train_loss": 0.07388268795609475,
-      "train_runtime": 21319.5402,
-      "train_samples_per_second": 3.002,
+      "epoch": 66.67,
+      "step": 100,
+      "total_flos": 42050959441920.0,
+      "train_loss": 0.3319866207242012,
+      "train_runtime": 4269.4137,
+      "train_samples_per_second": 2.998,
       "train_steps_per_second": 0.023
     }
   ],
-  "max_steps": 500,
-  "num_train_epochs": 500,
-  "total_flos": 210359990353920.0,
+  "max_steps": 100,
+  "num_train_epochs": 100,
+  "total_flos": 42050959441920.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2188a20782c485df4937a49e7b018f87feb047b94549b3e8907cdc9eff5c9801
-size 5563
+oid sha256:64e0c2d603edfa479a939173d26bb532d940f3ac3258651395bfbf4755154a1a
+size 5499