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_teknium_scaling_down_random_0.1
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # oh_teknium_scaling_down_random_0.1
 
-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_teknium_scaling_down_random_0.1 dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.3817
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.932475884244373,
+    "eval_loss": 0.38172662258148193,
+    "eval_runtime": 27.0168,
+    "eval_samples_per_second": 38.791,
+    "eval_steps_per_second": 0.629,
+    "total_flos": 190744866324480.0,
+    "train_loss": 0.3773096245631837,
+    "train_runtime": 4004.6609,
+    "train_samples_per_second": 14.911,
+    "train_steps_per_second": 0.028
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.932475884244373,
+    "eval_loss": 0.38172662258148193,
+    "eval_runtime": 27.0168,
+    "eval_samples_per_second": 38.791,
+    "eval_steps_per_second": 0.629
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.932475884244373,
+    "total_flos": 190744866324480.0,
+    "train_loss": 0.3773096245631837,
+    "train_runtime": 4004.6609,
+    "train_samples_per_second": 14.911,
+    "train_steps_per_second": 0.028
+}

trainer_state.json

@@ -0,0 +1,143 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.932475884244373,
+  "eval_steps": 500,
+  "global_step": 114,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2572347266881029,
+      "grad_norm": 1.4320514902529629,
+      "learning_rate": 5e-06,
+      "loss": 0.5408,
+      "step": 10
+    },
+    {
+      "epoch": 0.5144694533762058,
+      "grad_norm": 2.2238089511891426,
+      "learning_rate": 5e-06,
+      "loss": 0.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.7717041800643086,
+      "grad_norm": 0.7441671316236123,
+      "learning_rate": 5e-06,
+      "loss": 0.4329,
+      "step": 30
+    },
+    {
+      "epoch": 0.977491961414791,
+      "eval_loss": 0.41939377784729004,
+      "eval_runtime": 27.6464,
+      "eval_samples_per_second": 37.907,
+      "eval_steps_per_second": 0.615,
+      "step": 38
+    },
+    {
+      "epoch": 1.0289389067524115,
+      "grad_norm": 0.7629087732257928,
+      "learning_rate": 5e-06,
+      "loss": 0.418,
+      "step": 40
+    },
+    {
+      "epoch": 1.2861736334405145,
+      "grad_norm": 0.7572179893490683,
+      "learning_rate": 5e-06,
+      "loss": 0.3666,
+      "step": 50
+    },
+    {
+      "epoch": 1.5434083601286175,
+      "grad_norm": 0.6231742111626204,
+      "learning_rate": 5e-06,
+      "loss": 0.3579,
+      "step": 60
+    },
+    {
+      "epoch": 1.8006430868167203,
+      "grad_norm": 0.6946368492416183,
+      "learning_rate": 5e-06,
+      "loss": 0.3569,
+      "step": 70
+    },
+    {
+      "epoch": 1.9807073954983923,
+      "eval_loss": 0.38599085807800293,
+      "eval_runtime": 27.9317,
+      "eval_samples_per_second": 37.52,
+      "eval_steps_per_second": 0.609,
+      "step": 77
+    },
+    {
+      "epoch": 2.057877813504823,
+      "grad_norm": 0.8452067524392513,
+      "learning_rate": 5e-06,
+      "loss": 0.338,
+      "step": 80
+    },
+    {
+      "epoch": 2.315112540192926,
+      "grad_norm": 0.7194607593500192,
+      "learning_rate": 5e-06,
+      "loss": 0.3031,
+      "step": 90
+    },
+    {
+      "epoch": 2.572347266881029,
+      "grad_norm": 0.8382579339856899,
+      "learning_rate": 5e-06,
+      "loss": 0.2965,
+      "step": 100
+    },
+    {
+      "epoch": 2.829581993569132,
+      "grad_norm": 0.4969134197730987,
+      "learning_rate": 5e-06,
+      "loss": 0.2978,
+      "step": 110
+    },
+    {
+      "epoch": 2.932475884244373,
+      "eval_loss": 0.38172662258148193,
+      "eval_runtime": 25.7536,
+      "eval_samples_per_second": 40.693,
+      "eval_steps_per_second": 0.66,
+      "step": 114
+    },
+    {
+      "epoch": 2.932475884244373,
+      "step": 114,
+      "total_flos": 190744866324480.0,
+      "train_loss": 0.3773096245631837,
+      "train_runtime": 4004.6609,
+      "train_samples_per_second": 14.911,
+      "train_steps_per_second": 0.028
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 114,
+  "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": 190744866324480.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}