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_unstructured_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_unstructured dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0849
+- Loss: 0.0787
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.8828828828828827,
+    "eval_loss": 0.07866430282592773,
+    "eval_runtime": 92.8383,
+    "eval_samples_per_second": 15.769,
+    "eval_steps_per_second": 1.971,
+    "total_flos": 1.2688077065394586e+17,
+    "train_loss": 0.117483252286911,
+    "train_runtime": 3922.4935,
+    "train_samples_per_second": 5.432,
+    "train_steps_per_second": 0.021
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.8828828828828827,
+    "eval_loss": 0.07866430282592773,
+    "eval_runtime": 92.8383,
+    "eval_samples_per_second": 15.769,
+    "eval_steps_per_second": 1.971
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.8828828828828827,
+    "total_flos": 1.2688077065394586e+17,
+    "train_loss": 0.117483252286911,
+    "train_runtime": 3922.4935,
+    "train_samples_per_second": 5.432,
+    "train_steps_per_second": 0.021
+}

trainer_state.json

@@ -0,0 +1,171 @@
+{
+  "best_metric": 0.07866430282592773,
+  "best_model_checkpoint": "saves/Mistral-7B/lora/train_1/checkpoint-50",
+  "epoch": 2.8828828828828827,
+  "eval_steps": 10,
+  "global_step": 80,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.36036036036036034,
+      "grad_norm": 0.26596060395240784,
+      "learning_rate": 0.00028885859539033357,
+      "loss": 0.4784,
+      "step": 10
+    },
+    {
+      "epoch": 0.36036036036036034,
+      "eval_loss": 0.10740000009536743,
+      "eval_runtime": 92.81,
+      "eval_samples_per_second": 15.774,
+      "eval_steps_per_second": 1.972,
+      "step": 10
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "grad_norm": 0.3412328362464905,
+      "learning_rate": 0.00025708946018368484,
+      "loss": 0.0943,
+      "step": 20
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "eval_loss": 0.08951539546251297,
+      "eval_runtime": 92.8999,
+      "eval_samples_per_second": 15.759,
+      "eval_steps_per_second": 1.97,
+      "step": 20
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 0.26343515515327454,
+      "learning_rate": 0.0002094119649058735,
+      "loss": 0.0763,
+      "step": 30
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "eval_loss": 0.08366864919662476,
+      "eval_runtime": 92.931,
+      "eval_samples_per_second": 15.754,
+      "eval_steps_per_second": 1.969,
+      "step": 30
+    },
+    {
+      "epoch": 1.4414414414414414,
+      "grad_norm": 0.20433703064918518,
+      "learning_rate": 0.00015290869976577364,
+      "loss": 0.0674,
+      "step": 40
+    },
+    {
+      "epoch": 1.4414414414414414,
+      "eval_loss": 0.08040930330753326,
+      "eval_runtime": 92.8073,
+      "eval_samples_per_second": 15.775,
+      "eval_steps_per_second": 1.972,
+      "step": 40
+    },
+    {
+      "epoch": 1.8018018018018018,
+      "grad_norm": 0.23807112872600555,
+      "learning_rate": 9.597334127929346e-05,
+      "loss": 0.0622,
+      "step": 50
+    },
+    {
+      "epoch": 1.8018018018018018,
+      "eval_loss": 0.07866430282592773,
+      "eval_runtime": 92.8443,
+      "eval_samples_per_second": 15.768,
+      "eval_steps_per_second": 1.971,
+      "step": 50
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "grad_norm": 0.21819865703582764,
+      "learning_rate": 4.706375431968997e-05,
+      "loss": 0.0596,
+      "step": 60
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "eval_loss": 0.08040966093540192,
+      "eval_runtime": 92.8758,
+      "eval_samples_per_second": 15.763,
+      "eval_steps_per_second": 1.97,
+      "step": 60
+    },
+    {
+      "epoch": 2.5225225225225225,
+      "grad_norm": 0.23914609849452972,
+      "learning_rate": 1.3445558855078014e-05,
+      "loss": 0.0493,
+      "step": 70
+    },
+    {
+      "epoch": 2.5225225225225225,
+      "eval_loss": 0.08447360247373581,
+      "eval_runtime": 92.8784,
+      "eval_samples_per_second": 15.763,
+      "eval_steps_per_second": 1.97,
+      "step": 70
+    },
+    {
+      "epoch": 2.8828828828828827,
+      "grad_norm": 0.24320539832115173,
+      "learning_rate": 1.1280712436549378e-07,
+      "loss": 0.0523,
+      "step": 80
+    },
+    {
+      "epoch": 2.8828828828828827,
+      "eval_loss": 0.08488883823156357,
+      "eval_runtime": 92.8598,
+      "eval_samples_per_second": 15.766,
+      "eval_steps_per_second": 1.971,
+      "step": 80
+    },
+    {
+      "epoch": 2.8828828828828827,
+      "step": 80,
+      "total_flos": 1.2688077065394586e+17,
+      "train_loss": 0.117483252286911,
+      "train_runtime": 3922.4935,
+      "train_samples_per_second": 5.432,
+      "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": 1.2688077065394586e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}