End of training

README.md

@@ -2,6 +2,7 @@
 library_name: peft
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-v0.3
 model-index:
@@ -14,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral_ei_oc_structured_train
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the emollms_ei_oc_structured dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0838
+- Loss: 0.0778
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.5225225225225225,
+    "eval_loss": 0.07778054475784302,
+    "eval_runtime": 100.4856,
+    "eval_samples_per_second": 14.569,
+    "eval_steps_per_second": 1.821,
+    "total_flos": 1.2534445331801702e+17,
+    "train_loss": 0.12733433970383234,
+    "train_runtime": 3811.9913,
+    "train_samples_per_second": 5.589,
+    "train_steps_per_second": 0.021
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.5225225225225225,
+    "eval_loss": 0.07778054475784302,
+    "eval_runtime": 100.4856,
+    "eval_samples_per_second": 14.569,
+    "eval_steps_per_second": 1.821
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.5225225225225225,
+    "total_flos": 1.2534445331801702e+17,
+    "train_loss": 0.12733433970383234,
+    "train_runtime": 3811.9913,
+    "train_samples_per_second": 5.589,
+    "train_steps_per_second": 0.021
+}

trainer_state.json

@@ -0,0 +1,156 @@
+{
+  "best_metric": 0.07778054475784302,
+  "best_model_checkpoint": "saves/Mistral-7B/lora/train_1/checkpoint-40",
+  "epoch": 2.5225225225225225,
+  "eval_steps": 10,
+  "global_step": 70,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.36036036036036034,
+      "grad_norm": 2.3245933055877686,
+      "learning_rate": 0.00028885859539033357,
+      "loss": 0.4849,
+      "step": 10
+    },
+    {
+      "epoch": 0.36036036036036034,
+      "eval_loss": 0.1087382510304451,
+      "eval_runtime": 100.6667,
+      "eval_samples_per_second": 14.543,
+      "eval_steps_per_second": 1.818,
+      "step": 10
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "grad_norm": 0.36245808005332947,
+      "learning_rate": 0.00025708946018368484,
+      "loss": 0.0931,
+      "step": 20
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "eval_loss": 0.09484288841485977,
+      "eval_runtime": 100.7009,
+      "eval_samples_per_second": 14.538,
+      "eval_steps_per_second": 1.817,
+      "step": 20
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 0.280573308467865,
+      "learning_rate": 0.0002094119649058735,
+      "loss": 0.0758,
+      "step": 30
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "eval_loss": 0.08145211637020111,
+      "eval_runtime": 100.6686,
+      "eval_samples_per_second": 14.543,
+      "eval_steps_per_second": 1.818,
+      "step": 30
+    },
+    {
+      "epoch": 1.4414414414414414,
+      "grad_norm": 0.23253190517425537,
+      "learning_rate": 0.00015290869976577364,
+      "loss": 0.067,
+      "step": 40
+    },
+    {
+      "epoch": 1.4414414414414414,
+      "eval_loss": 0.07778054475784302,
+      "eval_runtime": 100.6925,
+      "eval_samples_per_second": 14.539,
+      "eval_steps_per_second": 1.817,
+      "step": 40
+    },
+    {
+      "epoch": 1.8018018018018018,
+      "grad_norm": 0.16911348700523376,
+      "learning_rate": 9.597334127929346e-05,
+      "loss": 0.0619,
+      "step": 50
+    },
+    {
+      "epoch": 1.8018018018018018,
+      "eval_loss": 0.07814132422208786,
+      "eval_runtime": 100.7056,
+      "eval_samples_per_second": 14.537,
+      "eval_steps_per_second": 1.817,
+      "step": 50
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "grad_norm": 0.21044674515724182,
+      "learning_rate": 4.706375431968997e-05,
+      "loss": 0.059,
+      "step": 60
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "eval_loss": 0.08178748935461044,
+      "eval_runtime": 100.6624,
+      "eval_samples_per_second": 14.544,
+      "eval_steps_per_second": 1.818,
+      "step": 60
+    },
+    {
+      "epoch": 2.5225225225225225,
+      "grad_norm": 0.21770545840263367,
+      "learning_rate": 1.3445558855078014e-05,
+      "loss": 0.0497,
+      "step": 70
+    },
+    {
+      "epoch": 2.5225225225225225,
+      "eval_loss": 0.0837995633482933,
+      "eval_runtime": 100.73,
+      "eval_samples_per_second": 14.534,
+      "eval_steps_per_second": 1.817,
+      "step": 70
+    },
+    {
+      "epoch": 2.5225225225225225,
+      "step": 70,
+      "total_flos": 1.2534445331801702e+17,
+      "train_loss": 0.12733433970383234,
+      "train_runtime": 3811.9913,
+      "train_samples_per_second": 5.589,
+      "train_steps_per_second": 0.021
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 81,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 10,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.2534445331801702e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}