Model save

README.md

@@ -0,0 +1,60 @@
+---
+license: llama3.1
+base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
+tags:
+- trl
+- sft
+- generated_from_trainer
+model-index:
+- name: llama3.1_8b_instruct_sft
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# llama3.1_8b_instruct_sft
+
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on an unknown dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-07
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 1024
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 8.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.44.0
+- Pytorch 2.4.0+cu121
+- Datasets 3.0.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.688888888888889,
+    "total_flos": 8467019857920.0,
+    "train_loss": 2.377727910876274,
+    "train_runtime": 2281.8745,
+    "train_samples": 2862,
+    "train_samples_per_second": 10.034,
+    "train_steps_per_second": 0.007
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.44.0"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.688888888888889,
+    "total_flos": 8467019857920.0,
+    "train_loss": 2.377727910876274,
+    "train_runtime": 2281.8745,
+    "train_samples": 2862,
+    "train_samples_per_second": 10.034,
+    "train_steps_per_second": 0.007
+}

trainer_state.json

@@ -0,0 +1,154 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.688888888888889,
+  "eval_steps": 100,
+  "global_step": 16,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.35555555555555557,
+      "grad_norm": 143.33721923828125,
+      "learning_rate": 2.5e-07,
+      "loss": 2.719,
+      "step": 1
+    },
+    {
+      "epoch": 0.7111111111111111,
+      "grad_norm": 144.41957092285156,
+      "learning_rate": 5e-07,
+      "loss": 2.738,
+      "step": 2
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 113.99625396728516,
+      "learning_rate": 4.937319780454559e-07,
+      "loss": 2.6899,
+      "step": 3
+    },
+    {
+      "epoch": 1.4222222222222223,
+      "grad_norm": 76.7990951538086,
+      "learning_rate": 4.752422169756047e-07,
+      "loss": 2.6816,
+      "step": 4
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 67.85514831542969,
+      "learning_rate": 4.4545787061700746e-07,
+      "loss": 2.5968,
+      "step": 5
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 67.69243621826172,
+      "learning_rate": 4.058724504646834e-07,
+      "loss": 2.5637,
+      "step": 6
+    },
+    {
+      "epoch": 2.488888888888889,
+      "grad_norm": 58.46062088012695,
+      "learning_rate": 3.584709347793895e-07,
+      "loss": 2.3581,
+      "step": 7
+    },
+    {
+      "epoch": 2.8444444444444446,
+      "grad_norm": 58.78388595581055,
+      "learning_rate": 3.056302334890786e-07,
+      "loss": 2.3439,
+      "step": 8
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 59.05506896972656,
+      "learning_rate": 2.5e-07,
+      "loss": 2.2871,
+      "step": 9
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 65.5489730834961,
+      "learning_rate": 1.9436976651092142e-07,
+      "loss": 2.2914,
+      "step": 10
+    },
+    {
+      "epoch": 3.911111111111111,
+      "grad_norm": 74.62954711914062,
+      "learning_rate": 1.4152906522061047e-07,
+      "loss": 2.2547,
+      "step": 11
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 76.41258239746094,
+      "learning_rate": 9.412754953531663e-08,
+      "loss": 2.2695,
+      "step": 12
+    },
+    {
+      "epoch": 4.622222222222222,
+      "grad_norm": 35.00369644165039,
+      "learning_rate": 5.454212938299255e-08,
+      "loss": 2.1083,
+      "step": 13
+    },
+    {
+      "epoch": 4.977777777777778,
+      "grad_norm": 66.7103271484375,
+      "learning_rate": 2.475778302439524e-08,
+      "loss": 2.0591,
+      "step": 14
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 59.597145080566406,
+      "learning_rate": 6.268021954544095e-09,
+      "loss": 2.0415,
+      "step": 15
+    },
+    {
+      "epoch": 5.688888888888889,
+      "grad_norm": 57.84925079345703,
+      "learning_rate": 0.0,
+      "loss": 2.0409,
+      "step": 16
+    },
+    {
+      "epoch": 5.688888888888889,
+      "step": 16,
+      "total_flos": 8467019857920.0,
+      "train_loss": 2.377727910876274,
+      "train_runtime": 2281.8745,
+      "train_samples_per_second": 10.034,
+      "train_steps_per_second": 0.007
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 16,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8467019857920.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}