End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Meta-Llama-3.1-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: stackexchange_law
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_law
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/stackexchange_law dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.9372
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9734513274336285,
+    "eval_loss": 0.9371946454048157,
+    "eval_runtime": 18.8461,
+    "eval_samples_per_second": 40.274,
+    "eval_steps_per_second": 0.637,
+    "total_flos": 140493748961280.0,
+    "train_loss": 0.9860022891135443,
+    "train_runtime": 3030.012,
+    "train_samples_per_second": 14.265,
+    "train_steps_per_second": 0.028
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9734513274336285,
+    "eval_loss": 0.9371946454048157,
+    "eval_runtime": 18.8461,
+    "eval_samples_per_second": 40.274,
+    "eval_steps_per_second": 0.637
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9734513274336285,
+    "total_flos": 140493748961280.0,
+    "train_loss": 0.9860022891135443,
+    "train_runtime": 3030.012,
+    "train_samples_per_second": 14.265,
+    "train_steps_per_second": 0.028
+}

trainer_state.json

@@ -0,0 +1,122 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9734513274336285,
+  "eval_steps": 500,
+  "global_step": 84,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.35398230088495575,
+      "grad_norm": 2.6903553349713225,
+      "learning_rate": 5e-06,
+      "loss": 1.1675,
+      "step": 10
+    },
+    {
+      "epoch": 0.7079646017699115,
+      "grad_norm": 3.4982970947971745,
+      "learning_rate": 5e-06,
+      "loss": 1.0528,
+      "step": 20
+    },
+    {
+      "epoch": 0.9911504424778761,
+      "eval_loss": 0.9985043406486511,
+      "eval_runtime": 19.6688,
+      "eval_samples_per_second": 38.589,
+      "eval_steps_per_second": 0.61,
+      "step": 28
+    },
+    {
+      "epoch": 1.0619469026548674,
+      "grad_norm": 1.4069236270193914,
+      "learning_rate": 5e-06,
+      "loss": 1.0801,
+      "step": 30
+    },
+    {
+      "epoch": 1.415929203539823,
+      "grad_norm": 1.5477898962865304,
+      "learning_rate": 5e-06,
+      "loss": 0.962,
+      "step": 40
+    },
+    {
+      "epoch": 1.7699115044247788,
+      "grad_norm": 1.5073318648745306,
+      "learning_rate": 5e-06,
+      "loss": 0.9408,
+      "step": 50
+    },
+    {
+      "epoch": 1.9823008849557522,
+      "eval_loss": 0.9531683921813965,
+      "eval_runtime": 19.302,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 0.622,
+      "step": 56
+    },
+    {
+      "epoch": 2.1238938053097347,
+      "grad_norm": 2.156406366483967,
+      "learning_rate": 5e-06,
+      "loss": 0.9837,
+      "step": 60
+    },
+    {
+      "epoch": 2.47787610619469,
+      "grad_norm": 1.340633178663602,
+      "learning_rate": 5e-06,
+      "loss": 0.8791,
+      "step": 70
+    },
+    {
+      "epoch": 2.831858407079646,
+      "grad_norm": 1.6870083465403531,
+      "learning_rate": 5e-06,
+      "loss": 0.8707,
+      "step": 80
+    },
+    {
+      "epoch": 2.9734513274336285,
+      "eval_loss": 0.9371946454048157,
+      "eval_runtime": 18.5828,
+      "eval_samples_per_second": 40.844,
+      "eval_steps_per_second": 0.646,
+      "step": 84
+    },
+    {
+      "epoch": 2.9734513274336285,
+      "step": 84,
+      "total_flos": 140493748961280.0,
+      "train_loss": 0.9860022891135443,
+      "train_runtime": 3030.012,
+      "train_samples_per_second": 14.265,
+      "train_steps_per_second": 0.028
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 84,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "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": 140493748961280.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}