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_v1-2_only_camel_chemistry
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # oh_v1-2_only_camel_chemistry
 
-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_v1-2_only_camel_chemistry dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6064
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9640914036996735,
+    "eval_loss": 0.6063705682754517,
+    "eval_runtime": 20.6052,
+    "eval_samples_per_second": 37.563,
+    "eval_steps_per_second": 0.631,
+    "total_flos": 140650783703040.0,
+    "train_loss": 0.6340561310450236,
+    "train_runtime": 3203.0166,
+    "train_samples_per_second": 13.762,
+    "train_steps_per_second": 0.026
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9640914036996735,
+    "eval_loss": 0.6063705682754517,
+    "eval_runtime": 20.6052,
+    "eval_samples_per_second": 37.563,
+    "eval_steps_per_second": 0.631
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9640914036996735,
+    "total_flos": 140650783703040.0,
+    "train_loss": 0.6340561310450236,
+    "train_runtime": 3203.0166,
+    "train_samples_per_second": 13.762,
+    "train_steps_per_second": 0.026
+}

trainer_state.json

@@ -0,0 +1,122 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9640914036996735,
+  "eval_steps": 500,
+  "global_step": 84,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3482045701849837,
+      "grad_norm": 2.8275081644658964,
+      "learning_rate": 5e-06,
+      "loss": 0.7569,
+      "step": 10
+    },
+    {
+      "epoch": 0.6964091403699674,
+      "grad_norm": 1.3641943960072136,
+      "learning_rate": 5e-06,
+      "loss": 0.6798,
+      "step": 20
+    },
+    {
+      "epoch": 0.9749727965179543,
+      "eval_loss": 0.6369664669036865,
+      "eval_runtime": 21.0013,
+      "eval_samples_per_second": 36.855,
+      "eval_steps_per_second": 0.619,
+      "step": 28
+    },
+    {
+      "epoch": 1.0642002176278564,
+      "grad_norm": 1.3194745810309563,
+      "learning_rate": 5e-06,
+      "loss": 0.7149,
+      "step": 30
+    },
+    {
+      "epoch": 1.41240478781284,
+      "grad_norm": 0.6203166856059423,
+      "learning_rate": 5e-06,
+      "loss": 0.6126,
+      "step": 40
+    },
+    {
+      "epoch": 1.7606093579978237,
+      "grad_norm": 0.5807474748224962,
+      "learning_rate": 5e-06,
+      "loss": 0.6059,
+      "step": 50
+    },
+    {
+      "epoch": 1.969532100108814,
+      "eval_loss": 0.6109879016876221,
+      "eval_runtime": 21.3562,
+      "eval_samples_per_second": 36.242,
+      "eval_steps_per_second": 0.609,
+      "step": 56
+    },
+    {
+      "epoch": 2.128400435255713,
+      "grad_norm": 0.8240309629325455,
+      "learning_rate": 5e-06,
+      "loss": 0.6248,
+      "step": 60
+    },
+    {
+      "epoch": 2.4766050054406965,
+      "grad_norm": 0.5773498698916745,
+      "learning_rate": 5e-06,
+      "loss": 0.5568,
+      "step": 70
+    },
+    {
+      "epoch": 2.82480957562568,
+      "grad_norm": 0.8304954778369694,
+      "learning_rate": 5e-06,
+      "loss": 0.552,
+      "step": 80
+    },
+    {
+      "epoch": 2.9640914036996735,
+      "eval_loss": 0.6063705682754517,
+      "eval_runtime": 19.6993,
+      "eval_samples_per_second": 39.291,
+      "eval_steps_per_second": 0.66,
+      "step": 84
+    },
+    {
+      "epoch": 2.9640914036996735,
+      "step": 84,
+      "total_flos": 140650783703040.0,
+      "train_loss": 0.6340561310450236,
+      "train_runtime": 3203.0166,
+      "train_samples_per_second": 13.762,
+      "train_steps_per_second": 0.026
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 84,
+  "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": 140650783703040.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}