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_math
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_math
 
-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_math dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6119
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9822485207100593,
+    "eval_loss": 0.6119207143783569,
+    "eval_runtime": 28.673,
+    "eval_samples_per_second": 39.689,
+    "eval_steps_per_second": 0.628,
+    "total_flos": 210845313269760.0,
+    "train_loss": 0.6075608276185536,
+    "train_runtime": 4413.1086,
+    "train_samples_per_second": 14.697,
+    "train_steps_per_second": 0.029
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9822485207100593,
+    "eval_loss": 0.6119207143783569,
+    "eval_runtime": 28.673,
+    "eval_samples_per_second": 39.689,
+    "eval_steps_per_second": 0.628
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9822485207100593,
+    "total_flos": 210845313269760.0,
+    "train_loss": 0.6075608276185536,
+    "train_runtime": 4413.1086,
+    "train_samples_per_second": 14.697,
+    "train_steps_per_second": 0.029
+}

trainer_state.json

@@ -0,0 +1,150 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9822485207100593,
+  "eval_steps": 500,
+  "global_step": 126,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.23668639053254437,
+      "grad_norm": 7.8563451765334875,
+      "learning_rate": 5e-06,
+      "loss": 0.7715,
+      "step": 10
+    },
+    {
+      "epoch": 0.47337278106508873,
+      "grad_norm": 1.2653404754348714,
+      "learning_rate": 5e-06,
+      "loss": 0.6752,
+      "step": 20
+    },
+    {
+      "epoch": 0.7100591715976331,
+      "grad_norm": 0.7722723244728256,
+      "learning_rate": 5e-06,
+      "loss": 0.6462,
+      "step": 30
+    },
+    {
+      "epoch": 0.9467455621301775,
+      "grad_norm": 0.5467401391418546,
+      "learning_rate": 5e-06,
+      "loss": 0.629,
+      "step": 40
+    },
+    {
+      "epoch": 0.9940828402366864,
+      "eval_loss": 0.6316158175468445,
+      "eval_runtime": 29.2463,
+      "eval_samples_per_second": 38.911,
+      "eval_steps_per_second": 0.615,
+      "step": 42
+    },
+    {
+      "epoch": 1.183431952662722,
+      "grad_norm": 0.6046791079019197,
+      "learning_rate": 5e-06,
+      "loss": 0.6475,
+      "step": 50
+    },
+    {
+      "epoch": 1.4201183431952662,
+      "grad_norm": 0.5148633963275753,
+      "learning_rate": 5e-06,
+      "loss": 0.5883,
+      "step": 60
+    },
+    {
+      "epoch": 1.6568047337278107,
+      "grad_norm": 0.5201018481905905,
+      "learning_rate": 5e-06,
+      "loss": 0.5813,
+      "step": 70
+    },
+    {
+      "epoch": 1.893491124260355,
+      "grad_norm": 0.876067375733763,
+      "learning_rate": 5e-06,
+      "loss": 0.5807,
+      "step": 80
+    },
+    {
+      "epoch": 1.9881656804733727,
+      "eval_loss": 0.6118831038475037,
+      "eval_runtime": 29.9679,
+      "eval_samples_per_second": 37.974,
+      "eval_steps_per_second": 0.601,
+      "step": 84
+    },
+    {
+      "epoch": 2.1301775147928996,
+      "grad_norm": 1.237055001310703,
+      "learning_rate": 5e-06,
+      "loss": 0.5993,
+      "step": 90
+    },
+    {
+      "epoch": 2.366863905325444,
+      "grad_norm": 0.657072150861852,
+      "learning_rate": 5e-06,
+      "loss": 0.5376,
+      "step": 100
+    },
+    {
+      "epoch": 2.603550295857988,
+      "grad_norm": 0.6298128100266783,
+      "learning_rate": 5e-06,
+      "loss": 0.5388,
+      "step": 110
+    },
+    {
+      "epoch": 2.8402366863905324,
+      "grad_norm": 0.6849566956882345,
+      "learning_rate": 5e-06,
+      "loss": 0.5369,
+      "step": 120
+    },
+    {
+      "epoch": 2.9822485207100593,
+      "eval_loss": 0.6119207143783569,
+      "eval_runtime": 28.1349,
+      "eval_samples_per_second": 40.448,
+      "eval_steps_per_second": 0.64,
+      "step": 126
+    },
+    {
+      "epoch": 2.9822485207100593,
+      "step": 126,
+      "total_flos": 210845313269760.0,
+      "train_loss": 0.6075608276185536,
+      "train_runtime": 4413.1086,
+      "train_samples_per_second": 14.697,
+      "train_steps_per_second": 0.029
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 126,
+  "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": 210845313269760.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}