End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Llama-3.1-8B-Instruct
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: sky-t1-original-llama-instruct
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # sky-t1-original-llama-instruct
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the mlfoundations-dev/sky_t1_original dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.4424
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.4424304664134979,
+    "eval_runtime": 11.6183,
+    "eval_samples_per_second": 69.545,
+    "eval_steps_per_second": 1.119,
+    "total_flos": 150543972433920.0,
+    "train_loss": 0.4572353945838081,
+    "train_runtime": 2587.8961,
+    "train_samples_per_second": 17.787,
+    "train_steps_per_second": 0.035
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.4424304664134979,
+    "eval_runtime": 11.6183,
+    "eval_samples_per_second": 69.545,
+    "eval_steps_per_second": 1.119
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 150543972433920.0,
+    "train_loss": 0.4572353945838081,
+    "train_runtime": 2587.8961,
+    "train_samples_per_second": 17.787,
+    "train_steps_per_second": 0.035
+}

trainer_state.json

@@ -0,0 +1,129 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 90,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 1.1328805578866286,
+      "learning_rate": 5e-06,
+      "loss": 0.6261,
+      "step": 10
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.5655437748332994,
+      "learning_rate": 5e-06,
+      "loss": 0.5229,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.4255185947425879,
+      "learning_rate": 5e-06,
+      "loss": 0.4914,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.48239657282829285,
+      "eval_runtime": 11.6513,
+      "eval_samples_per_second": 69.349,
+      "eval_steps_per_second": 1.116,
+      "step": 30
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4517651743600907,
+      "learning_rate": 5e-06,
+      "loss": 0.4487,
+      "step": 40
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.3649131361498104,
+      "learning_rate": 5e-06,
+      "loss": 0.438,
+      "step": 50
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.3830225718161711,
+      "learning_rate": 5e-06,
+      "loss": 0.4301,
+      "step": 60
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.45320141315460205,
+      "eval_runtime": 11.8663,
+      "eval_samples_per_second": 68.092,
+      "eval_steps_per_second": 1.096,
+      "step": 60
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.47711530562220955,
+      "learning_rate": 5e-06,
+      "loss": 0.3922,
+      "step": 70
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.47041140445746077,
+      "learning_rate": 5e-06,
+      "loss": 0.3841,
+      "step": 80
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.43944682743936286,
+      "learning_rate": 5e-06,
+      "loss": 0.3817,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.4424304664134979,
+      "eval_runtime": 11.6409,
+      "eval_samples_per_second": 69.41,
+      "eval_steps_per_second": 1.117,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "step": 90,
+      "total_flos": 150543972433920.0,
+      "train_loss": 0.4572353945838081,
+      "train_runtime": 2587.8961,
+      "train_samples_per_second": 17.787,
+      "train_steps_per_second": 0.035
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 90,
+  "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": 150543972433920.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}