End of training

README.md

@@ -3,6 +3,7 @@ license: other
 base_model: MathSymbol/BasicSFT_1.8_Pretrain_Lightning
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: WordProblem
@@ -14,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # WordProblem
 
-This model is a fine-tuned version of [MathSymbol/BasicSFT_1.8_Pretrain_Lightning](https://huggingface.co/MathSymbol/BasicSFT_1.8_Pretrain_Lightning) on an unknown dataset.
+This model is a fine-tuned version of [MathSymbol/BasicSFT_1.8_Pretrain_Lightning](https://huggingface.co/MathSymbol/BasicSFT_1.8_Pretrain_Lightning) on the WordProblems_SFT dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.1677
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 0.9999725884707108,
+    "eval_loss": 0.1676628440618515,
+    "eval_runtime": 79.4786,
+    "eval_samples_per_second": 37.482,
+    "eval_steps_per_second": 9.374,
+    "total_flos": 3.6976201313039155e+17,
+    "train_loss": 0.19690959160788019,
+    "train_runtime": 20589.0374,
+    "train_samples_per_second": 7.087,
+    "train_steps_per_second": 0.443
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 0.9999725884707108,
+    "eval_loss": 0.1676628440618515,
+    "eval_runtime": 79.4786,
+    "eval_samples_per_second": 37.482,
+    "eval_steps_per_second": 9.374
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9999725884707108,
+    "total_flos": 3.6976201313039155e+17,
+    "train_loss": 0.19690959160788019,
+    "train_runtime": 20589.0374,
+    "train_samples_per_second": 7.087,
+    "train_steps_per_second": 0.443
+}

trainer_log.jsonl

@@ -35,3 +35,4 @@
 {"current_steps": 9000, "total_steps": 9120, "loss": 0.168, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3904986054812396e-08, "epoch": 0.9868150544118857, "percentage": 98.68, "elapsed_time": "5:36:30", "remaining_time": "0:04:29"}
 {"current_steps": 9000, "total_steps": 9120, "loss": null, "eval_loss": 0.1676628440618515, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.9868150544118857, "percentage": 98.68, "elapsed_time": "5:36:30", "remaining_time": "0:04:29"}
 {"current_steps": 9120, "total_steps": 9120, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.9999725884707108, "percentage": 100.0, "elapsed_time": "5:43:04", "remaining_time": "0:00:00"}
+{"current_steps": 745, "total_steps": 745, "loss": null, "eval_loss": 0.1676628440618515, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.9999725884707108, "percentage": 100.0, "elapsed_time": "5:44:46", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,288 @@
+{
+  "best_metric": 0.1676628440618515,
+  "best_model_checkpoint": "saves/Qwen1.5-1.8B/WordProblem/checkpoint-9000",
+  "epoch": 0.9999725884707108,
+  "eval_steps": 1500,
+  "global_step": 9120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03289383514706285,
+      "grad_norm": 4.8125,
+      "learning_rate": 3e-05,
+      "loss": 0.7143,
+      "step": 300
+    },
+    {
+      "epoch": 0.0657876702941257,
+      "grad_norm": 3.5,
+      "learning_rate": 4.998339850669331e-05,
+      "loss": 0.2219,
+      "step": 600
+    },
+    {
+      "epoch": 0.09868150544118856,
+      "grad_norm": 4.40625,
+      "learning_rate": 4.9734816848192624e-05,
+      "loss": 0.2074,
+      "step": 900
+    },
+    {
+      "epoch": 0.1315753405882514,
+      "grad_norm": 4.03125,
+      "learning_rate": 4.9190839785031474e-05,
+      "loss": 0.1906,
+      "step": 1200
+    },
+    {
+      "epoch": 0.16446917573531428,
+      "grad_norm": 3.78125,
+      "learning_rate": 4.835796376008569e-05,
+      "loss": 0.1923,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16446917573531428,
+      "eval_loss": 0.1849033087491989,
+      "eval_runtime": 79.6037,
+      "eval_samples_per_second": 37.423,
+      "eval_steps_per_second": 9.359,
+      "step": 1500
+    },
+    {
+      "epoch": 0.19736301088237712,
+      "grad_norm": 4.15625,
+      "learning_rate": 4.7246135390382216e-05,
+      "loss": 0.1839,
+      "step": 1800
+    },
+    {
+      "epoch": 0.23025684602944,
+      "grad_norm": 4.5,
+      "learning_rate": 4.586863267968384e-05,
+      "loss": 0.1938,
+      "step": 2100
+    },
+    {
+      "epoch": 0.2631506811765028,
+      "grad_norm": 3.1875,
+      "learning_rate": 4.4241906446007296e-05,
+      "loss": 0.1863,
+      "step": 2400
+    },
+    {
+      "epoch": 0.2960445163235657,
+      "grad_norm": 2.71875,
+      "learning_rate": 4.238538385782601e-05,
+      "loss": 0.1797,
+      "step": 2700
+    },
+    {
+      "epoch": 0.32893835147062855,
+      "grad_norm": 3.65625,
+      "learning_rate": 4.032123642522486e-05,
+      "loss": 0.176,
+      "step": 3000
+    },
+    {
+      "epoch": 0.32893835147062855,
+      "eval_loss": 0.1760552078485489,
+      "eval_runtime": 79.6333,
+      "eval_samples_per_second": 37.409,
+      "eval_steps_per_second": 9.355,
+      "step": 3000
+    },
+    {
+      "epoch": 0.3618321866176914,
+      "grad_norm": 4.0,
+      "learning_rate": 3.8074115216771435e-05,
+      "loss": 0.1791,
+      "step": 3300
+    },
+    {
+      "epoch": 0.39472602176475424,
+      "grad_norm": 3.90625,
+      "learning_rate": 3.567085646427478e-05,
+      "loss": 0.1808,
+      "step": 3600
+    },
+    {
+      "epoch": 0.4276198569118171,
+      "grad_norm": 3.421875,
+      "learning_rate": 3.3140161071244915e-05,
+      "loss": 0.1805,
+      "step": 3900
+    },
+    {
+      "epoch": 0.46051369205888,
+      "grad_norm": 2.640625,
+      "learning_rate": 3.05122518525215e-05,
+      "loss": 0.1738,
+      "step": 4200
+    },
+    {
+      "epoch": 0.49340752720594283,
+      "grad_norm": 4.5,
+      "learning_rate": 2.781851259848554e-05,
+      "loss": 0.1736,
+      "step": 4500
+    },
+    {
+      "epoch": 0.49340752720594283,
+      "eval_loss": 0.17090687155723572,
+      "eval_runtime": 79.6329,
+      "eval_samples_per_second": 37.409,
+      "eval_steps_per_second": 9.355,
+      "step": 4500
+    },
+    {
+      "epoch": 0.5263013623530056,
+      "grad_norm": 3.578125,
+      "learning_rate": 2.509111327432736e-05,
+      "loss": 0.1709,
+      "step": 4800
+    },
+    {
+      "epoch": 0.5591951975000685,
+      "grad_norm": 3.515625,
+      "learning_rate": 2.236262583042668e-05,
+      "loss": 0.1775,
+      "step": 5100
+    },
+    {
+      "epoch": 0.5920890326471314,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.966563521202681e-05,
+      "loss": 0.1759,
+      "step": 5400
+    },
+    {
+      "epoch": 0.6249828677941942,
+      "grad_norm": 3.640625,
+      "learning_rate": 1.7032350213717874e-05,
+      "loss": 0.1754,
+      "step": 5700
+    },
+    {
+      "epoch": 0.6578767029412571,
+      "grad_norm": 3.578125,
+      "learning_rate": 1.4494218826096939e-05,
+      "loss": 0.1688,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6578767029412571,
+      "eval_loss": 0.16823573410511017,
+      "eval_runtime": 79.6163,
+      "eval_samples_per_second": 37.417,
+      "eval_steps_per_second": 9.357,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6907705380883199,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.2081552668325321e-05,
+      "loss": 0.1707,
+      "step": 6300
+    },
+    {
+      "epoch": 0.7236643732353828,
+      "grad_norm": 4.0625,
+      "learning_rate": 9.82316499179518e-06,
+      "loss": 0.171,
+      "step": 6600
+    },
+    {
+      "epoch": 0.7565582083824457,
+      "grad_norm": 3.984375,
+      "learning_rate": 7.74602657804425e-06,
+      "loss": 0.1702,
+      "step": 6900
+    },
+    {
+      "epoch": 0.7894520435295085,
+      "grad_norm": 3.421875,
+      "learning_rate": 5.874943640356082e-06,
+      "loss": 0.1718,
+      "step": 7200
+    },
+    {
+      "epoch": 0.8223458786765714,
+      "grad_norm": 4.1875,
+      "learning_rate": 4.232261575703861e-06,
+      "loss": 0.1689,
+      "step": 7500
+    },
+    {
+      "epoch": 0.8223458786765714,
+      "eval_loss": 0.16773280501365662,
+      "eval_runtime": 79.6198,
+      "eval_samples_per_second": 37.415,
+      "eval_steps_per_second": 9.357,
+      "step": 7500
+    },
+    {
+      "epoch": 0.8552397138236342,
+      "grad_norm": 2.703125,
+      "learning_rate": 2.83759810497852e-06,
+      "loss": 0.1692,
+      "step": 7800
+    },
+    {
+      "epoch": 0.8881335489706971,
+      "grad_norm": 4.0625,
+      "learning_rate": 1.70760898847247e-06,
+      "loss": 0.1787,
+      "step": 8100
+    },
+    {
+      "epoch": 0.92102738411776,
+      "grad_norm": 4.71875,
+      "learning_rate": 8.557891145603042e-07,
+      "loss": 0.1733,
+      "step": 8400
+    },
+    {
+      "epoch": 0.9539212192648228,
+      "grad_norm": 3.6875,
+      "learning_rate": 2.923113370737779e-07,
+      "loss": 0.1741,
+      "step": 8700
+    },
+    {
+      "epoch": 0.9868150544118857,
+      "grad_norm": 4.3125,
+      "learning_rate": 2.3904986054812396e-08,
+      "loss": 0.168,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9868150544118857,
+      "eval_loss": 0.1676628440618515,
+      "eval_runtime": 79.7085,
+      "eval_samples_per_second": 37.374,
+      "eval_steps_per_second": 9.347,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9999725884707108,
+      "step": 9120,
+      "total_flos": 3.6976201313039155e+17,
+      "train_loss": 0.19690959160788019,
+      "train_runtime": 20589.0374,
+      "train_samples_per_second": 7.087,
+      "train_steps_per_second": 0.443
+    }
+  ],
+  "logging_steps": 300,
+  "max_steps": 9120,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 3000,
+  "total_flos": 3.6976201313039155e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}