Model save

README.md

@@ -0,0 +1,69 @@
+---
+library_name: peft
+license: apache-2.0
+base_model: mistralai/Mistral-7B-v0.3
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: mistral7b-milora-summarization-11-v1
+  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. -->
+
+# mistral7b-milora-summarization-11-v1
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0168
+
+## 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: 0.0002
+- train_batch_size: 14
+- eval_batch_size: 14
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 224
+- total_eval_batch_size: 112
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.3739        | 0.9965 | 142  | 2.0168          |
+
+
+### Framework versions
+
+- PEFT 0.13.3.dev0
+- Transformers 4.46.3
+- Pytorch 2.3.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9964912280701754,
+    "total_flos": 6.954534912540017e+17,
+    "train_loss": 1.4842075613183034,
+    "train_runtime": 638.3404,
+    "train_samples": 129221,
+    "train_samples_per_second": 49.937,
+    "train_steps_per_second": 0.222
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9964912280701754,
+    "total_flos": 6.954534912540017e+17,
+    "train_loss": 1.4842075613183034,
+    "train_runtime": 638.3404,
+    "train_samples": 129221,
+    "train_samples_per_second": 49.937,
+    "train_steps_per_second": 0.222
+}

trainer_state.json

@@ -0,0 +1,253 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9964912280701754,
+  "eval_steps": 500,
+  "global_step": 142,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007017543859649123,
+      "grad_norm": 3.2127857208251953,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 2.0873,
+      "step": 1
+    },
+    {
+      "epoch": 0.03508771929824561,
+      "grad_norm": 2.7192869186401367,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 2.0735,
+      "step": 5
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 1.4391745328903198,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 1.989,
+      "step": 10
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 1.263310432434082,
+      "learning_rate": 0.0002,
+      "loss": 1.9059,
+      "step": 15
+    },
+    {
+      "epoch": 0.14035087719298245,
+      "grad_norm": 1.1776515245437622,
+      "learning_rate": 0.00019923607874151032,
+      "loss": 1.7644,
+      "step": 20
+    },
+    {
+      "epoch": 0.17543859649122806,
+      "grad_norm": 0.7352864146232605,
+      "learning_rate": 0.00019695598647982468,
+      "loss": 1.6067,
+      "step": 25
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.5063140392303467,
+      "learning_rate": 0.00019319455943394347,
+      "loss": 1.5369,
+      "step": 30
+    },
+    {
+      "epoch": 0.24561403508771928,
+      "grad_norm": 0.4345757067203522,
+      "learning_rate": 0.00018800926628551886,
+      "loss": 1.4784,
+      "step": 35
+    },
+    {
+      "epoch": 0.2807017543859649,
+      "grad_norm": 0.29390987753868103,
+      "learning_rate": 0.00018147933014790244,
+      "loss": 1.4507,
+      "step": 40
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 0.30241647362709045,
+      "learning_rate": 0.0001737045181617364,
+      "loss": 1.441,
+      "step": 45
+    },
+    {
+      "epoch": 0.3508771929824561,
+      "grad_norm": 0.32600656151771545,
+      "learning_rate": 0.00016480361721016054,
+      "loss": 1.4313,
+      "step": 50
+    },
+    {
+      "epoch": 0.38596491228070173,
+      "grad_norm": 0.3142453730106354,
+      "learning_rate": 0.00015491261904230727,
+      "loss": 1.4181,
+      "step": 55
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.26200494170188904,
+      "learning_rate": 0.0001441826425335387,
+      "loss": 1.4184,
+      "step": 60
+    },
+    {
+      "epoch": 0.45614035087719296,
+      "grad_norm": 0.23935051262378693,
+      "learning_rate": 0.00013277762482701767,
+      "loss": 1.4003,
+      "step": 65
+    },
+    {
+      "epoch": 0.49122807017543857,
+      "grad_norm": 0.2776546776294708,
+      "learning_rate": 0.00012087181663233354,
+      "loss": 1.3959,
+      "step": 70
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.26008203625679016,
+      "learning_rate": 0.00010864711994907458,
+      "loss": 1.3973,
+      "step": 75
+    },
+    {
+      "epoch": 0.5614035087719298,
+      "grad_norm": 0.2285134345293045,
+      "learning_rate": 9.629030889073949e-05,
+      "loss": 1.3859,
+      "step": 80
+    },
+    {
+      "epoch": 0.5964912280701754,
+      "grad_norm": 0.2977670431137085,
+      "learning_rate": 8.399017607042025e-05,
+      "loss": 1.3798,
+      "step": 85
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.26052919030189514,
+      "learning_rate": 7.193464814699073e-05,
+      "loss": 1.378,
+      "step": 90
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.2686891257762909,
+      "learning_rate": 6.0307914601711305e-05,
+      "loss": 1.3713,
+      "step": 95
+    },
+    {
+      "epoch": 0.7017543859649122,
+      "grad_norm": 0.24683193862438202,
+      "learning_rate": 4.928761361302269e-05,
+      "loss": 1.3818,
+      "step": 100
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.26356831192970276,
+      "learning_rate": 3.904211802492922e-05,
+      "loss": 1.3778,
+      "step": 105
+    },
+    {
+      "epoch": 0.7719298245614035,
+      "grad_norm": 0.25175535678863525,
+      "learning_rate": 2.9727962875101e-05,
+      "loss": 1.3744,
+      "step": 110
+    },
+    {
+      "epoch": 0.8070175438596491,
+      "grad_norm": 0.27307969331741333,
+      "learning_rate": 2.1487453786014512e-05,
+      "loss": 1.3687,
+      "step": 115
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.22797873616218567,
+      "learning_rate": 1.4446492759148411e-05,
+      "loss": 1.3829,
+      "step": 120
+    },
+    {
+      "epoch": 0.8771929824561403,
+      "grad_norm": 0.24017217755317688,
+      "learning_rate": 8.712654590675085e-06,
+      "loss": 1.3692,
+      "step": 125
+    },
+    {
+      "epoch": 0.9122807017543859,
+      "grad_norm": 0.22941164672374725,
+      "learning_rate": 4.37354329798726e-06,
+      "loss": 1.3739,
+      "step": 130
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.21575631201267242,
+      "learning_rate": 1.4954536682736719e-06,
+      "loss": 1.3737,
+      "step": 135
+    },
+    {
+      "epoch": 0.9824561403508771,
+      "grad_norm": 0.2508339583873749,
+      "learning_rate": 1.2235837857387246e-07,
+      "loss": 1.3739,
+      "step": 140
+    },
+    {
+      "epoch": 0.9964912280701754,
+      "eval_loss": 2.0167758464813232,
+      "eval_runtime": 0.7,
+      "eval_samples_per_second": 34.285,
+      "eval_steps_per_second": 1.429,
+      "step": 142
+    },
+    {
+      "epoch": 0.9964912280701754,
+      "step": 142,
+      "total_flos": 6.954534912540017e+17,
+      "train_loss": 1.4842075613183034,
+      "train_runtime": 638.3404,
+      "train_samples_per_second": 49.937,
+      "train_steps_per_second": 0.222
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 142,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.954534912540017e+17,
+  "train_batch_size": 14,
+  "trial_name": null,
+  "trial_params": null
+}