End of training

README.md

@@ -4,6 +4,7 @@ license: mit
 base_model: microsoft/Phi-4-mini-instruct
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: Phi-4-mini-instruct_sft_sg_values
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Phi-4-mini-instruct_sft_sg_values
 
-This model is a fine-tuned version of [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) on an unknown dataset.
+This model is a fine-tuned version of [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) on the sft_sg_values dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3880
+- Loss: 2.3848
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 0.9996261682242991,
+    "eval_loss": 2.384812116622925,
+    "eval_runtime": 18.1789,
+    "eval_samples_per_second": 27.504,
+    "eval_steps_per_second": 13.752,
+    "total_flos": 4.450466167504896e+16,
+    "train_loss": 3.11931433781101,
+    "train_runtime": 1360.775,
+    "train_samples_per_second": 7.863,
+    "train_steps_per_second": 0.983
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 0.9996261682242991,
+    "eval_loss": 2.384812116622925,
+    "eval_runtime": 18.1789,
+    "eval_samples_per_second": 27.504,
+    "eval_steps_per_second": 13.752
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9996261682242991,
+    "total_flos": 4.450466167504896e+16,
+    "train_loss": 3.11931433781101,
+    "train_runtime": 1360.775,
+    "train_samples_per_second": 7.863,
+    "train_steps_per_second": 0.983
+}

trainer_state.json

@@ -0,0 +1,216 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9996261682242991,
+  "eval_steps": 250,
+  "global_step": 1337,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05009345794392523,
+      "grad_norm": 3.513535261154175,
+      "learning_rate": 4.925373134328357e-07,
+      "loss": 4.4864,
+      "step": 67
+    },
+    {
+      "epoch": 0.10018691588785046,
+      "grad_norm": 4.232038974761963,
+      "learning_rate": 9.925373134328357e-07,
+      "loss": 4.4815,
+      "step": 134
+    },
+    {
+      "epoch": 0.1502803738317757,
+      "grad_norm": 3.0024209022521973,
+      "learning_rate": 9.451371571072319e-07,
+      "loss": 4.3614,
+      "step": 201
+    },
+    {
+      "epoch": 0.18691588785046728,
+      "eval_loss": 4.165477752685547,
+      "eval_runtime": 18.4458,
+      "eval_samples_per_second": 27.107,
+      "eval_steps_per_second": 13.553,
+      "step": 250
+    },
+    {
+      "epoch": 0.20037383177570092,
+      "grad_norm": 4.35124397277832,
+      "learning_rate": 8.894430590191188e-07,
+      "loss": 4.1954,
+      "step": 268
+    },
+    {
+      "epoch": 0.2504672897196262,
+      "grad_norm": 4.761136531829834,
+      "learning_rate": 8.337489609310058e-07,
+      "loss": 3.9922,
+      "step": 335
+    },
+    {
+      "epoch": 0.3005607476635514,
+      "grad_norm": 5.12410831451416,
+      "learning_rate": 7.780548628428927e-07,
+      "loss": 3.6946,
+      "step": 402
+    },
+    {
+      "epoch": 0.3506542056074766,
+      "grad_norm": 4.370376110076904,
+      "learning_rate": 7.223607647547797e-07,
+      "loss": 3.3914,
+      "step": 469
+    },
+    {
+      "epoch": 0.37383177570093457,
+      "eval_loss": 3.1697182655334473,
+      "eval_runtime": 18.4011,
+      "eval_samples_per_second": 27.172,
+      "eval_steps_per_second": 13.586,
+      "step": 500
+    },
+    {
+      "epoch": 0.40074766355140184,
+      "grad_norm": 3.3430747985839844,
+      "learning_rate": 6.666666666666666e-07,
+      "loss": 3.154,
+      "step": 536
+    },
+    {
+      "epoch": 0.4508411214953271,
+      "grad_norm": 4.1145734786987305,
+      "learning_rate": 6.109725685785536e-07,
+      "loss": 2.9973,
+      "step": 603
+    },
+    {
+      "epoch": 0.5009345794392523,
+      "grad_norm": 3.4338088035583496,
+      "learning_rate": 5.552784704904405e-07,
+      "loss": 2.815,
+      "step": 670
+    },
+    {
+      "epoch": 0.5510280373831775,
+      "grad_norm": 3.3768129348754883,
+      "learning_rate": 4.995843724023275e-07,
+      "loss": 2.684,
+      "step": 737
+    },
+    {
+      "epoch": 0.5607476635514018,
+      "eval_loss": 2.6262145042419434,
+      "eval_runtime": 18.4215,
+      "eval_samples_per_second": 27.142,
+      "eval_steps_per_second": 13.571,
+      "step": 750
+    },
+    {
+      "epoch": 0.6011214953271028,
+      "grad_norm": 2.7695705890655518,
+      "learning_rate": 4.438902743142144e-07,
+      "loss": 2.6152,
+      "step": 804
+    },
+    {
+      "epoch": 0.6512149532710281,
+      "grad_norm": 2.307760715484619,
+      "learning_rate": 3.881961762261014e-07,
+      "loss": 2.5592,
+      "step": 871
+    },
+    {
+      "epoch": 0.7013084112149532,
+      "grad_norm": 2.533203125,
+      "learning_rate": 3.3250207813798835e-07,
+      "loss": 2.5099,
+      "step": 938
+    },
+    {
+      "epoch": 0.7476635514018691,
+      "eval_loss": 2.444204807281494,
+      "eval_runtime": 18.3848,
+      "eval_samples_per_second": 27.196,
+      "eval_steps_per_second": 13.598,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7514018691588785,
+      "grad_norm": 1.7434587478637695,
+      "learning_rate": 2.7680798004987534e-07,
+      "loss": 2.4359,
+      "step": 1005
+    },
+    {
+      "epoch": 0.8014953271028037,
+      "grad_norm": 2.9051830768585205,
+      "learning_rate": 2.2111388196176226e-07,
+      "loss": 2.4193,
+      "step": 1072
+    },
+    {
+      "epoch": 0.851588785046729,
+      "grad_norm": 2.5767340660095215,
+      "learning_rate": 1.6541978387364923e-07,
+      "loss": 2.4261,
+      "step": 1139
+    },
+    {
+      "epoch": 0.9016822429906542,
+      "grad_norm": 1.5444096326828003,
+      "learning_rate": 1.0972568578553615e-07,
+      "loss": 2.3706,
+      "step": 1206
+    },
+    {
+      "epoch": 0.9345794392523364,
+      "eval_loss": 2.3879618644714355,
+      "eval_runtime": 18.3821,
+      "eval_samples_per_second": 27.2,
+      "eval_steps_per_second": 13.6,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9517757009345794,
+      "grad_norm": 2.0951781272888184,
+      "learning_rate": 5.403158769742311e-08,
+      "loss": 2.3796,
+      "step": 1273
+    },
+    {
+      "epoch": 0.9996261682242991,
+      "step": 1337,
+      "total_flos": 4.450466167504896e+16,
+      "train_loss": 3.11931433781101,
+      "train_runtime": 1360.775,
+      "train_samples_per_second": 7.863,
+      "train_steps_per_second": 0.983
+    }
+  ],
+  "logging_steps": 67,
+  "max_steps": 1337,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.450466167504896e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}