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_metamath_40k
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # oh_v3-1_only_metamath_40k
 
-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_metamath_40k dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2824
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.891566265060241,
+    "eval_loss": 0.2824314832687378,
+    "eval_runtime": 74.7328,
+    "eval_samples_per_second": 5.607,
+    "eval_steps_per_second": 0.361,
+    "total_flos": 75324331130880.0,
+    "train_loss": 0.33777744240230984,
+    "train_runtime": 12667.2664,
+    "train_samples_per_second": 1.885,
+    "train_steps_per_second": 0.004
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.891566265060241,
+    "eval_loss": 0.2824314832687378,
+    "eval_runtime": 74.7328,
+    "eval_samples_per_second": 5.607,
+    "eval_steps_per_second": 0.361
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.891566265060241,
+    "total_flos": 75324331130880.0,
+    "train_loss": 0.33777744240230984,
+    "train_runtime": 12667.2664,
+    "train_samples_per_second": 1.885,
+    "train_steps_per_second": 0.004
+}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.891566265060241,
+  "eval_steps": 500,
+  "global_step": 45,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.642570281124498,
+      "grad_norm": 3.008957806586854,
+      "learning_rate": 5e-06,
+      "loss": 0.4671,
+      "step": 10
+    },
+    {
+      "epoch": 0.963855421686747,
+      "eval_loss": 0.33529338240623474,
+      "eval_runtime": 74.8385,
+      "eval_samples_per_second": 5.599,
+      "eval_steps_per_second": 0.361,
+      "step": 15
+    },
+    {
+      "epoch": 1.285140562248996,
+      "grad_norm": 1.4278343000102331,
+      "learning_rate": 5e-06,
+      "loss": 0.3359,
+      "step": 20
+    },
+    {
+      "epoch": 1.927710843373494,
+      "grad_norm": 0.9256547137281664,
+      "learning_rate": 5e-06,
+      "loss": 0.3019,
+      "step": 30
+    },
+    {
+      "epoch": 1.9919678714859437,
+      "eval_loss": 0.2970140278339386,
+      "eval_runtime": 74.8069,
+      "eval_samples_per_second": 5.601,
+      "eval_steps_per_second": 0.361,
+      "step": 31
+    },
+    {
+      "epoch": 2.570281124497992,
+      "grad_norm": 1.8739185721481015,
+      "learning_rate": 5e-06,
+      "loss": 0.2815,
+      "step": 40
+    },
+    {
+      "epoch": 2.891566265060241,
+      "eval_loss": 0.2824314832687378,
+      "eval_runtime": 74.7028,
+      "eval_samples_per_second": 5.609,
+      "eval_steps_per_second": 0.361,
+      "step": 45
+    },
+    {
+      "epoch": 2.891566265060241,
+      "step": 45,
+      "total_flos": 75324331130880.0,
+      "train_loss": 0.33777744240230984,
+      "train_runtime": 12667.2664,
+      "train_samples_per_second": 1.885,
+      "train_steps_per_second": 0.004
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 45,
+  "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": 75324331130880.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}