End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Llama-3.1-8B-Instruct
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: llama_3.1_large_crafting_sft_success
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama_3.1_large_crafting_sft_success
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the None dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the identity and the large_crafting_sft_success datasets.
 It achieves the following results on the evaluation set:
-- Loss: 0.4033
+- Loss: 0.3760
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 0.9936628643852978,
+    "eval_loss": 0.3760089576244354,
+    "eval_runtime": 10.4117,
+    "eval_samples_per_second": 37.17,
+    "eval_steps_per_second": 4.706,
+    "total_flos": 93124196827136.0,
+    "train_loss": 0.5249049436478388,
+    "train_runtime": 2087.7125,
+    "train_samples_per_second": 9.069,
+    "train_steps_per_second": 0.07
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 0.9936628643852978,
+    "eval_loss": 0.3760089576244354,
+    "eval_runtime": 10.4117,
+    "eval_samples_per_second": 37.17,
+    "eval_steps_per_second": 4.706
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9936628643852978,
+    "total_flos": 93124196827136.0,
+    "train_loss": 0.5249049436478388,
+    "train_runtime": 2087.7125,
+    "train_samples_per_second": 9.069,
+    "train_steps_per_second": 0.07
+}

trainer_state.json

@@ -0,0 +1,156 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9936628643852978,
+  "eval_steps": 50,
+  "global_step": 147,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.06759611322348964,
+      "grad_norm": 8.075992652271083,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.009,
+      "step": 10
+    },
+    {
+      "epoch": 0.13519222644697929,
+      "grad_norm": 4.741723636049252,
+      "learning_rate": 9.964639423366442e-06,
+      "loss": 0.6126,
+      "step": 20
+    },
+    {
+      "epoch": 0.20278833967046894,
+      "grad_norm": 3.792115901987726,
+      "learning_rate": 9.68474862499881e-06,
+      "loss": 0.5753,
+      "step": 30
+    },
+    {
+      "epoch": 0.27038445289395857,
+      "grad_norm": 3.9179679148908293,
+      "learning_rate": 9.140746393556853e-06,
+      "loss": 0.5607,
+      "step": 40
+    },
+    {
+      "epoch": 0.33798056611744826,
+      "grad_norm": 3.794130346554311,
+      "learning_rate": 8.363301868506264e-06,
+      "loss": 0.5283,
+      "step": 50
+    },
+    {
+      "epoch": 0.33798056611744826,
+      "eval_loss": 0.4638153314590454,
+      "eval_runtime": 10.5122,
+      "eval_samples_per_second": 36.814,
+      "eval_steps_per_second": 4.661,
+      "step": 50
+    },
+    {
+      "epoch": 0.4055766793409379,
+      "grad_norm": 3.311164882131189,
+      "learning_rate": 7.396244933600285e-06,
+      "loss": 0.5108,
+      "step": 60
+    },
+    {
+      "epoch": 0.47317279256442757,
+      "grad_norm": 3.532561734971584,
+      "learning_rate": 6.294095225512604e-06,
+      "loss": 0.4907,
+      "step": 70
+    },
+    {
+      "epoch": 0.5407689057879171,
+      "grad_norm": 3.0570492611506714,
+      "learning_rate": 5.118988487730537e-06,
+      "loss": 0.4703,
+      "step": 80
+    },
+    {
+      "epoch": 0.6083650190114068,
+      "grad_norm": 3.454847700200177,
+      "learning_rate": 3.937173552235117e-06,
+      "loss": 0.4738,
+      "step": 90
+    },
+    {
+      "epoch": 0.6759611322348965,
+      "grad_norm": 3.276095441165215,
+      "learning_rate": 2.8152774381532033e-06,
+      "loss": 0.4551,
+      "step": 100
+    },
+    {
+      "epoch": 0.6759611322348965,
+      "eval_loss": 0.4032706022262573,
+      "eval_runtime": 10.5071,
+      "eval_samples_per_second": 36.832,
+      "eval_steps_per_second": 4.664,
+      "step": 100
+    },
+    {
+      "epoch": 0.7435572454583862,
+      "grad_norm": 3.0153350107336223,
+      "learning_rate": 1.8165491294045596e-06,
+      "loss": 0.4633,
+      "step": 110
+    },
+    {
+      "epoch": 0.8111533586818758,
+      "grad_norm": 2.7044255298211644,
+      "learning_rate": 9.972937953781985e-07,
+      "loss": 0.4219,
+      "step": 120
+    },
+    {
+      "epoch": 0.8787494719053655,
+      "grad_norm": 3.627016753602074,
+      "learning_rate": 4.036984820916723e-07,
+      "loss": 0.4158,
+      "step": 130
+    },
+    {
+      "epoch": 0.9463455851288551,
+      "grad_norm": 2.8450276262646037,
+      "learning_rate": 6.922823140906754e-08,
+      "loss": 0.4425,
+      "step": 140
+    },
+    {
+      "epoch": 0.9936628643852978,
+      "step": 147,
+      "total_flos": 93124196827136.0,
+      "train_loss": 0.5249049436478388,
+      "train_runtime": 2087.7125,
+      "train_samples_per_second": 9.069,
+      "train_steps_per_second": 0.07
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 147,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "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": 93124196827136.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}