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_biology
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # stackexchange_biology
 
-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_biology dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.9231
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.948356807511737,
+    "eval_loss": 0.9231159687042236,
+    "eval_runtime": 29.7017,
+    "eval_samples_per_second": 24.106,
+    "eval_steps_per_second": 0.404,
+    "total_flos": 130443525488640.0,
+    "train_loss": 0.9576788682204026,
+    "train_runtime": 4829.447,
+    "train_samples_per_second": 8.445,
+    "train_steps_per_second": 0.016
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.948356807511737,
+    "eval_loss": 0.9231159687042236,
+    "eval_runtime": 29.7017,
+    "eval_samples_per_second": 24.106,
+    "eval_steps_per_second": 0.404
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.948356807511737,
+    "total_flos": 130443525488640.0,
+    "train_loss": 0.9576788682204026,
+    "train_runtime": 4829.447,
+    "train_samples_per_second": 8.445,
+    "train_steps_per_second": 0.016
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.948356807511737,
+  "eval_steps": 500,
+  "global_step": 78,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3755868544600939,
+      "grad_norm": 7.420562088875575,
+      "learning_rate": 5e-06,
+      "loss": 1.1267,
+      "step": 10
+    },
+    {
+      "epoch": 0.7511737089201878,
+      "grad_norm": 1.9326575993897155,
+      "learning_rate": 5e-06,
+      "loss": 1.0283,
+      "step": 20
+    },
+    {
+      "epoch": 0.9765258215962441,
+      "eval_loss": 0.9831027984619141,
+      "eval_runtime": 29.9012,
+      "eval_samples_per_second": 23.946,
+      "eval_steps_per_second": 0.401,
+      "step": 26
+    },
+    {
+      "epoch": 1.136150234741784,
+      "grad_norm": 1.1316893276757012,
+      "learning_rate": 5e-06,
+      "loss": 1.039,
+      "step": 30
+    },
+    {
+      "epoch": 1.511737089201878,
+      "grad_norm": 0.9870625157508757,
+      "learning_rate": 5e-06,
+      "loss": 0.9304,
+      "step": 40
+    },
+    {
+      "epoch": 1.887323943661972,
+      "grad_norm": 1.1888711767104414,
+      "learning_rate": 5e-06,
+      "loss": 0.914,
+      "step": 50
+    },
+    {
+      "epoch": 1.9624413145539905,
+      "eval_loss": 0.9372476935386658,
+      "eval_runtime": 30.023,
+      "eval_samples_per_second": 23.848,
+      "eval_steps_per_second": 0.4,
+      "step": 52
+    },
+    {
+      "epoch": 2.272300469483568,
+      "grad_norm": 1.4024339606578018,
+      "learning_rate": 5e-06,
+      "loss": 0.9248,
+      "step": 60
+    },
+    {
+      "epoch": 2.647887323943662,
+      "grad_norm": 1.366121091039955,
+      "learning_rate": 5e-06,
+      "loss": 0.8388,
+      "step": 70
+    },
+    {
+      "epoch": 2.948356807511737,
+      "eval_loss": 0.9231159687042236,
+      "eval_runtime": 29.1774,
+      "eval_samples_per_second": 24.54,
+      "eval_steps_per_second": 0.411,
+      "step": 78
+    },
+    {
+      "epoch": 2.948356807511737,
+      "step": 78,
+      "total_flos": 130443525488640.0,
+      "train_loss": 0.9576788682204026,
+      "train_runtime": 4829.447,
+      "train_samples_per_second": 8.445,
+      "train_steps_per_second": 0.016
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 78,
+  "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": 130443525488640.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}