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-pissa-classification-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-pissa-classification-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: 1.9625
+
+## 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: 12
+- eval_batch_size: 12
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 192
+- total_eval_batch_size: 96
+- 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 |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.7055        | 0.9945 | 91   | 1.9625          |
+
+
+### Framework versions
+
+- PEFT 0.13.2
+- 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": 0.994535519125683,
+    "total_flos": 3.8200965830436454e+17,
+    "train_loss": 1.8459819070585481,
+    "train_runtime": 299.5258,
+    "train_samples": 92634,
+    "train_samples_per_second": 58.409,
+    "train_steps_per_second": 0.304
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.994535519125683,
+    "total_flos": 3.8200965830436454e+17,
+    "train_loss": 1.8459819070585481,
+    "train_runtime": 299.5258,
+    "train_samples": 92634,
+    "train_samples_per_second": 58.409,
+    "train_steps_per_second": 0.304
+}

trainer_state.json

@@ -0,0 +1,183 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.994535519125683,
+  "eval_steps": 500,
+  "global_step": 91,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01092896174863388,
+      "grad_norm": 9.203123092651367,
+      "learning_rate": 2e-05,
+      "loss": 2.582,
+      "step": 1
+    },
+    {
+      "epoch": 0.0546448087431694,
+      "grad_norm": 5.399723052978516,
+      "learning_rate": 0.0001,
+      "loss": 2.511,
+      "step": 5
+    },
+    {
+      "epoch": 0.1092896174863388,
+      "grad_norm": 2.4733173847198486,
+      "learning_rate": 0.0002,
+      "loss": 2.3222,
+      "step": 10
+    },
+    {
+      "epoch": 0.16393442622950818,
+      "grad_norm": 2.0453178882598877,
+      "learning_rate": 0.00019812553106273847,
+      "loss": 2.1174,
+      "step": 15
+    },
+    {
+      "epoch": 0.2185792349726776,
+      "grad_norm": 1.9896700382232666,
+      "learning_rate": 0.00019257239692688907,
+      "loss": 1.9656,
+      "step": 20
+    },
+    {
+      "epoch": 0.273224043715847,
+      "grad_norm": 0.9077722430229187,
+      "learning_rate": 0.00018354878114129367,
+      "loss": 1.8583,
+      "step": 25
+    },
+    {
+      "epoch": 0.32786885245901637,
+      "grad_norm": 0.7602087259292603,
+      "learning_rate": 0.00017139297345578994,
+      "loss": 1.7976,
+      "step": 30
+    },
+    {
+      "epoch": 0.3825136612021858,
+      "grad_norm": 0.8192912340164185,
+      "learning_rate": 0.00015656068754865387,
+      "loss": 1.7772,
+      "step": 35
+    },
+    {
+      "epoch": 0.4371584699453552,
+      "grad_norm": 0.6041179895401001,
+      "learning_rate": 0.0001396079766039157,
+      "loss": 1.7488,
+      "step": 40
+    },
+    {
+      "epoch": 0.4918032786885246,
+      "grad_norm": 0.5910283327102661,
+      "learning_rate": 0.0001211703872229411,
+      "loss": 1.7323,
+      "step": 45
+    },
+    {
+      "epoch": 0.546448087431694,
+      "grad_norm": 0.6845403909683228,
+      "learning_rate": 0.00010193913317718244,
+      "loss": 1.7316,
+      "step": 50
+    },
+    {
+      "epoch": 0.6010928961748634,
+      "grad_norm": 0.6234318614006042,
+      "learning_rate": 8.263518223330697e-05,
+      "loss": 1.7174,
+      "step": 55
+    },
+    {
+      "epoch": 0.6557377049180327,
+      "grad_norm": 0.5730881094932556,
+      "learning_rate": 6.398222751952899e-05,
+      "loss": 1.7068,
+      "step": 60
+    },
+    {
+      "epoch": 0.7103825136612022,
+      "grad_norm": 0.5278472304344177,
+      "learning_rate": 4.66795567198309e-05,
+      "loss": 1.7131,
+      "step": 65
+    },
+    {
+      "epoch": 0.7650273224043715,
+      "grad_norm": 0.519974946975708,
+      "learning_rate": 3.137583621312665e-05,
+      "loss": 1.71,
+      "step": 70
+    },
+    {
+      "epoch": 0.819672131147541,
+      "grad_norm": 0.5470304489135742,
+      "learning_rate": 1.864479297370325e-05,
+      "loss": 1.7065,
+      "step": 75
+    },
+    {
+      "epoch": 0.8743169398907104,
+      "grad_norm": 0.5100380182266235,
+      "learning_rate": 8.963705903385345e-06,
+      "loss": 1.7043,
+      "step": 80
+    },
+    {
+      "epoch": 0.9289617486338798,
+      "grad_norm": 0.45625337958335876,
+      "learning_rate": 2.6955129420176196e-06,
+      "loss": 1.6982,
+      "step": 85
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "grad_norm": 0.5059428215026855,
+      "learning_rate": 7.520474957699586e-08,
+      "loss": 1.7055,
+      "step": 90
+    },
+    {
+      "epoch": 0.994535519125683,
+      "eval_loss": 1.9624824523925781,
+      "eval_runtime": 0.5691,
+      "eval_samples_per_second": 7.028,
+      "eval_steps_per_second": 1.757,
+      "step": 91
+    },
+    {
+      "epoch": 0.994535519125683,
+      "step": 91,
+      "total_flos": 3.8200965830436454e+17,
+      "train_loss": 1.8459819070585481,
+      "train_runtime": 299.5258,
+      "train_samples_per_second": 58.409,
+      "train_steps_per_second": 0.304
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 91,
+  "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": 3.8200965830436454e+17,
+  "train_batch_size": 12,
+  "trial_name": null,
+  "trial_params": null
+}