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_dsp
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_dsp
 
-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_dsp dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.8000
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9822485207100593,
+    "eval_loss": 0.7999942302703857,
+    "eval_runtime": 28.6817,
+    "eval_samples_per_second": 39.642,
+    "eval_steps_per_second": 0.628,
+    "total_flos": 210845313269760.0,
+    "train_loss": 0.8213065730200874,
+    "train_runtime": 4456.5245,
+    "train_samples_per_second": 14.53,
+    "train_steps_per_second": 0.028
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9822485207100593,
+    "eval_loss": 0.7999942302703857,
+    "eval_runtime": 28.6817,
+    "eval_samples_per_second": 39.642,
+    "eval_steps_per_second": 0.628
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9822485207100593,
+    "total_flos": 210845313269760.0,
+    "train_loss": 0.8213065730200874,
+    "train_runtime": 4456.5245,
+    "train_samples_per_second": 14.53,
+    "train_steps_per_second": 0.028
+}

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": 1.9313669247614162,
+      "learning_rate": 5e-06,
+      "loss": 1.018,
+      "step": 10
+    },
+    {
+      "epoch": 0.47337278106508873,
+      "grad_norm": 1.379427316659985,
+      "learning_rate": 5e-06,
+      "loss": 0.9191,
+      "step": 20
+    },
+    {
+      "epoch": 0.7100591715976331,
+      "grad_norm": 1.6079229642198343,
+      "learning_rate": 5e-06,
+      "loss": 0.8677,
+      "step": 30
+    },
+    {
+      "epoch": 0.9467455621301775,
+      "grad_norm": 0.734643863022887,
+      "learning_rate": 5e-06,
+      "loss": 0.8493,
+      "step": 40
+    },
+    {
+      "epoch": 0.9940828402366864,
+      "eval_loss": 0.8392977118492126,
+      "eval_runtime": 30.1115,
+      "eval_samples_per_second": 37.76,
+      "eval_steps_per_second": 0.598,
+      "step": 42
+    },
+    {
+      "epoch": 1.183431952662722,
+      "grad_norm": 1.005967406304035,
+      "learning_rate": 5e-06,
+      "loss": 0.8734,
+      "step": 50
+    },
+    {
+      "epoch": 1.4201183431952662,
+      "grad_norm": 0.9370549861399187,
+      "learning_rate": 5e-06,
+      "loss": 0.7941,
+      "step": 60
+    },
+    {
+      "epoch": 1.6568047337278107,
+      "grad_norm": 0.7056985035855818,
+      "learning_rate": 5e-06,
+      "loss": 0.787,
+      "step": 70
+    },
+    {
+      "epoch": 1.893491124260355,
+      "grad_norm": 1.2176250359078529,
+      "learning_rate": 5e-06,
+      "loss": 0.7843,
+      "step": 80
+    },
+    {
+      "epoch": 1.9881656804733727,
+      "eval_loss": 0.8069051504135132,
+      "eval_runtime": 29.6227,
+      "eval_samples_per_second": 38.383,
+      "eval_steps_per_second": 0.608,
+      "step": 84
+    },
+    {
+      "epoch": 2.1301775147928996,
+      "grad_norm": 1.0149726634752274,
+      "learning_rate": 5e-06,
+      "loss": 0.8153,
+      "step": 90
+    },
+    {
+      "epoch": 2.366863905325444,
+      "grad_norm": 1.1966616271031878,
+      "learning_rate": 5e-06,
+      "loss": 0.7379,
+      "step": 100
+    },
+    {
+      "epoch": 2.603550295857988,
+      "grad_norm": 1.28942414214322,
+      "learning_rate": 5e-06,
+      "loss": 0.7355,
+      "step": 110
+    },
+    {
+      "epoch": 2.8402366863905324,
+      "grad_norm": 0.6282115192314033,
+      "learning_rate": 5e-06,
+      "loss": 0.7323,
+      "step": 120
+    },
+    {
+      "epoch": 2.9822485207100593,
+      "eval_loss": 0.7999942302703857,
+      "eval_runtime": 27.42,
+      "eval_samples_per_second": 41.466,
+      "eval_steps_per_second": 0.656,
+      "step": 126
+    },
+    {
+      "epoch": 2.9822485207100593,
+      "step": 126,
+      "total_flos": 210845313269760.0,
+      "train_loss": 0.8213065730200874,
+      "train_runtime": 4456.5245,
+      "train_samples_per_second": 14.53,
+      "train_steps_per_second": 0.028
+    }
+  ],
+  "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
+}