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: oh_v3-1_only_airoboros
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # oh_v3-1_only_airoboros
 
-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/oh_v3-1_only_airoboros dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.7652
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.765191376209259,
+    "eval_runtime": 10.6628,
+    "eval_samples_per_second": 40.421,
+    "eval_steps_per_second": 0.656,
+    "total_flos": 80192408125440.0,
+    "train_loss": 0.7525163888931274,
+    "train_runtime": 1791.1975,
+    "train_samples_per_second": 13.694,
+    "train_steps_per_second": 0.027
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 0.765191376209259,
+    "eval_runtime": 10.6628,
+    "eval_samples_per_second": 40.421,
+    "eval_steps_per_second": 0.656
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "total_flos": 80192408125440.0,
+    "train_loss": 0.7525163888931274,
+    "train_runtime": 1791.1975,
+    "train_samples_per_second": 13.694,
+    "train_steps_per_second": 0.027
+}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 48,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.625,
+      "grad_norm": 3.5597619707423953,
+      "learning_rate": 5e-06,
+      "loss": 0.8943,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.8237837553024292,
+      "eval_runtime": 11.794,
+      "eval_samples_per_second": 36.544,
+      "eval_steps_per_second": 0.594,
+      "step": 16
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.8460541987937592,
+      "learning_rate": 5e-06,
+      "loss": 0.7853,
+      "step": 20
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 1.0407485286288733,
+      "learning_rate": 5e-06,
+      "loss": 0.7293,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.7811707258224487,
+      "eval_runtime": 12.1038,
+      "eval_samples_per_second": 35.609,
+      "eval_steps_per_second": 0.578,
+      "step": 32
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 1.5701597782418673,
+      "learning_rate": 5e-06,
+      "loss": 0.677,
+      "step": 40
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.765191376209259,
+      "eval_runtime": 11.6755,
+      "eval_samples_per_second": 36.915,
+      "eval_steps_per_second": 0.6,
+      "step": 48
+    },
+    {
+      "epoch": 3.0,
+      "step": 48,
+      "total_flos": 80192408125440.0,
+      "train_loss": 0.7525163888931274,
+      "train_runtime": 1791.1975,
+      "train_samples_per_second": 13.694,
+      "train_steps_per_second": 0.027
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 48,
+  "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": 80192408125440.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}