End of training

README.md

@@ -6,6 +6,7 @@ metrics:
 - accuracy
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: llama2_7b_test_4
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama2_7b_test_4
 
-This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the alpaca_en_demo dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.6859
 - Accuracy: 0.6973

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 0.08888888888888889,
+    "eval_accuracy": 0.6973038730142216,
+    "eval_loss": 1.685868740081787,
+    "eval_runtime": 21.0785,
+    "eval_samples_per_second": 4.744,
+    "eval_steps_per_second": 4.744,
+    "total_flos": 591688209530880.0,
+    "train_loss": 1.8846651792526246,
+    "train_runtime": 97.4531,
+    "train_samples_per_second": 0.821,
+    "train_steps_per_second": 0.103
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.08888888888888889,
+    "eval_accuracy": 0.6973038730142216,
+    "eval_loss": 1.685868740081787,
+    "eval_runtime": 21.0785,
+    "eval_samples_per_second": 4.744,
+    "eval_steps_per_second": 4.744
+}

runs/Aug01_14-28-05_ip-10-192-10-179/events.out.tfevents.1722522664.ip-10-192-10-179.23427.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d76346f1091b8a822cf24489f696d89d53bb1e4a13ecf7d52deedd2ed3fff55
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.08888888888888889,
+    "total_flos": 591688209530880.0,
+    "train_loss": 1.8846651792526246,
+    "train_runtime": 97.4531,
+    "train_samples_per_second": 0.821,
+    "train_steps_per_second": 0.103
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 1.685868740081787,
+  "best_model_checkpoint": "saves/llama2_7b_test_4/checkpoint-8",
+  "epoch": 0.08888888888888889,
+  "eval_steps": 4,
+  "global_step": 10,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008888888888888889,
+      "grad_norm": 0.9656311869621277,
+      "learning_rate": 0.0001,
+      "loss": 1.469,
+      "step": 1
+    },
+    {
+      "epoch": 0.017777777777777778,
+      "grad_norm": 2.022883653640747,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 2.9751,
+      "step": 2
+    },
+    {
+      "epoch": 0.02666666666666667,
+      "grad_norm": 1.2307357788085938,
+      "learning_rate": 8.83022221559489e-05,
+      "loss": 1.8521,
+      "step": 3
+    },
+    {
+      "epoch": 0.035555555555555556,
+      "grad_norm": 2.503842830657959,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 2.5627,
+      "step": 4
+    },
+    {
+      "epoch": 0.035555555555555556,
+      "eval_accuracy": 0.6939260670748238,
+      "eval_loss": 1.9984581470489502,
+      "eval_runtime": 20.9951,
+      "eval_samples_per_second": 4.763,
+      "eval_steps_per_second": 4.763,
+      "step": 4
+    },
+    {
+      "epoch": 0.044444444444444446,
+      "grad_norm": 1.6151976585388184,
+      "learning_rate": 5.868240888334653e-05,
+      "loss": 1.7733,
+      "step": 5
+    },
+    {
+      "epoch": 0.05333333333333334,
+      "grad_norm": 1.3629400730133057,
+      "learning_rate": 4.131759111665349e-05,
+      "loss": 1.5353,
+      "step": 6
+    },
+    {
+      "epoch": 0.06222222222222222,
+      "grad_norm": 2.5036630630493164,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 1.3856,
+      "step": 7
+    },
+    {
+      "epoch": 0.07111111111111111,
+      "grad_norm": 2.502861499786377,
+      "learning_rate": 1.1697777844051105e-05,
+      "loss": 2.2846,
+      "step": 8
+    },
+    {
+      "epoch": 0.07111111111111111,
+      "eval_accuracy": 0.6973038730142216,
+      "eval_loss": 1.685868740081787,
+      "eval_runtime": 21.1016,
+      "eval_samples_per_second": 4.739,
+      "eval_steps_per_second": 4.739,
+      "step": 8
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 2.4686763286590576,
+      "learning_rate": 3.0153689607045845e-06,
+      "loss": 1.6419,
+      "step": 9
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "grad_norm": 1.3143010139465332,
+      "learning_rate": 0.0,
+      "loss": 1.367,
+      "step": 10
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "step": 10,
+      "total_flos": 591688209530880.0,
+      "train_loss": 1.8846651792526246,
+      "train_runtime": 97.4531,
+      "train_samples_per_second": 0.821,
+      "train_steps_per_second": 0.103
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 10,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 4,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 591688209530880.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}