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_mathematica
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_mathematica
 
-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_mathematica dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.7114
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.7114487886428833,
+    "eval_runtime": 26.5423,
+    "eval_samples_per_second": 40.614,
+    "eval_steps_per_second": 0.64,
+    "total_flos": 200795089797120.0,
+    "train_loss": 0.7261142373085022,
+    "train_runtime": 4198.995,
+    "train_samples_per_second": 14.629,
+    "train_steps_per_second": 0.029
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.7114487886428833,
+    "eval_runtime": 26.5423,
+    "eval_samples_per_second": 40.614,
+    "eval_steps_per_second": 0.64
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 200795089797120.0,
+    "train_loss": 0.7261142373085022,
+    "train_runtime": 4198.995,
+    "train_samples_per_second": 14.629,
+    "train_steps_per_second": 0.029
+}

trainer_state.json

@@ -0,0 +1,150 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.25,
+      "grad_norm": 10.623923460343729,
+      "learning_rate": 5e-06,
+      "loss": 0.9372,
+      "step": 10
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.793037918373031,
+      "learning_rate": 5e-06,
+      "loss": 0.8283,
+      "step": 20
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 3.7845451356616455,
+      "learning_rate": 5e-06,
+      "loss": 0.7811,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9164583944436338,
+      "learning_rate": 5e-06,
+      "loss": 0.7595,
+      "step": 40
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7516233921051025,
+      "eval_runtime": 27.7388,
+      "eval_samples_per_second": 38.862,
+      "eval_steps_per_second": 0.613,
+      "step": 40
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 1.0460327419425848,
+      "learning_rate": 5e-06,
+      "loss": 0.7169,
+      "step": 50
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.134555380206765,
+      "learning_rate": 5e-06,
+      "loss": 0.7105,
+      "step": 60
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.834786450946659,
+      "learning_rate": 5e-06,
+      "loss": 0.6992,
+      "step": 70
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8259020499128051,
+      "learning_rate": 5e-06,
+      "loss": 0.6946,
+      "step": 80
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.7181093096733093,
+      "eval_runtime": 27.9425,
+      "eval_samples_per_second": 38.579,
+      "eval_steps_per_second": 0.608,
+      "step": 80
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 0.6752448107919039,
+      "learning_rate": 5e-06,
+      "loss": 0.651,
+      "step": 90
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.763737793787202,
+      "learning_rate": 5e-06,
+      "loss": 0.6462,
+      "step": 100
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 1.0609101015318403,
+      "learning_rate": 5e-06,
+      "loss": 0.6444,
+      "step": 110
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.8435640146675477,
+      "learning_rate": 5e-06,
+      "loss": 0.6445,
+      "step": 120
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.7114487886428833,
+      "eval_runtime": 26.8405,
+      "eval_samples_per_second": 40.163,
+      "eval_steps_per_second": 0.633,
+      "step": 120
+    },
+    {
+      "epoch": 3.0,
+      "step": 120,
+      "total_flos": 200795089797120.0,
+      "train_loss": 0.7261142373085022,
+      "train_runtime": 4198.995,
+      "train_samples_per_second": 14.629,
+      "train_steps_per_second": 0.029
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 120,
+  "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": 200795089797120.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}