初始化项目，由ModelHub XC社区提供模型

Model: alakxender/dhivehi-gpt2-base Source: Original Platform
2026-07-10 10:52:12 +08:00
commit b06778f831
29 changed files with 300910 additions and 0 deletions
--- a/checkpoint-100/trainer_state.json
+++ b/checkpoint-100/trainer_state.json
@@ -0,0 +1,111 @@
+{
+  "best_metric": 1.382002830505371,
+  "best_model_checkpoint": "dhivehi-gpt2-base/checkpoint-100",
+  "epoch": 2.4390243902439024,
+  "eval_steps": 100,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.24390243902439024,
+      "grad_norm": 83.27034759521484,
+      "learning_rate": 2.5000000000000004e-07,
+      "loss": 2.4436,
+      "step": 10
+    },
+    {
+      "epoch": 0.4878048780487805,
+      "grad_norm": 60.58613967895508,
+      "learning_rate": 7.5e-07,
+      "loss": 2.321,
+      "step": 20
+    },
+    {
+      "epoch": 0.7317073170731707,
+      "grad_norm": 29.228893280029297,
+      "learning_rate": 1.25e-06,
+      "loss": 2.0685,
+      "step": 30
+    },
+    {
+      "epoch": 0.975609756097561,
+      "grad_norm": 18.66552734375,
+      "learning_rate": 1.7500000000000002e-06,
+      "loss": 1.8913,
+      "step": 40
+    },
+    {
+      "epoch": 1.2195121951219512,
+      "grad_norm": 10.73972225189209,
+      "learning_rate": 2.25e-06,
+      "loss": 1.7331,
+      "step": 50
+    },
+    {
+      "epoch": 1.4634146341463414,
+      "grad_norm": 6.437626361846924,
+      "learning_rate": 2.7500000000000004e-06,
+      "loss": 1.6317,
+      "step": 60
+    },
+    {
+      "epoch": 1.7073170731707317,
+      "grad_norm": 3.9424028396606445,
+      "learning_rate": 3.2500000000000002e-06,
+      "loss": 1.5445,
+      "step": 70
+    },
+    {
+      "epoch": 1.951219512195122,
+      "grad_norm": 2.4789206981658936,
+      "learning_rate": 3.75e-06,
+      "loss": 1.4638,
+      "step": 80
+    },
+    {
+      "epoch": 2.1951219512195124,
+      "grad_norm": 1.7527034282684326,
+      "learning_rate": 4.250000000000001e-06,
+      "loss": 1.3972,
+      "step": 90
+    },
+    {
+      "epoch": 2.4390243902439024,
+      "grad_norm": 1.9903717041015625,
+      "learning_rate": 4.75e-06,
+      "loss": 1.3272,
+      "step": 100
+    },
+    {
+      "epoch": 2.4390243902439024,
+      "eval_loss": 1.382002830505371,
+      "eval_runtime": 5.1108,
+      "eval_samples_per_second": 56.742,
+      "eval_steps_per_second": 1.957,
+      "step": 100
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 123,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3322589478912000.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}