Model save

README.md

@@ -0,0 +1,58 @@
+---
+license: apache-2.0
+base_model: mistralai/Mistral-7B-v0.1
+tags:
+- generated_from_trainer
+model-index:
+- name: Mistral-7B-v0.1-dpo-2k
+  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. -->
+
+# Mistral-7B-v0.1-dpo-2k
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) 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: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.43.3
+- Pytorch 2.2.2+cu121
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.992,
+    "total_flos": 0.0,
+    "train_loss": 0.8068127228367713,
+    "train_runtime": 741.4049,
+    "train_samples": 2000,
+    "train_samples_per_second": 2.698,
+    "train_steps_per_second": 0.084
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.43.3"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.992,
+    "total_flos": 0.0,
+    "train_loss": 0.8068127228367713,
+    "train_runtime": 741.4049,
+    "train_samples": 2000,
+    "train_samples_per_second": 2.698,
+    "train_steps_per_second": 0.084
+}

trainer_state.json

@@ -0,0 +1,147 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.992,
+  "eval_steps": 200,
+  "global_step": 62,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.016,
+      "grad_norm": 110.59263434727151,
+      "learning_rate": 7.142857142857142e-08,
+      "logits/generated": -2.6108503341674805,
+      "logits/real": -2.2121129035949707,
+      "logps/generated": -118.9744873046875,
+      "logps/real": -150.31668090820312,
+      "loss": 0.8477,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 100.26623043984584,
+      "learning_rate": 4.727272727272727e-07,
+      "logits/generated": -2.5122885704040527,
+      "logits/real": -2.559802770614624,
+      "logps/generated": -132.923095703125,
+      "logps/real": -140.05096435546875,
+      "loss": 0.8088,
+      "rewards/accuracies": 0.4027777910232544,
+      "rewards/generated": 0.229824960231781,
+      "rewards/margins": 0.03599211201071739,
+      "rewards/real": 0.2658171057701111,
+      "step": 10
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 84.13081729415536,
+      "learning_rate": 3.818181818181818e-07,
+      "logits/generated": -2.5297508239746094,
+      "logits/real": -2.5737991333007812,
+      "logps/generated": -121.31373596191406,
+      "logps/real": -120.0510025024414,
+      "loss": 0.8275,
+      "rewards/accuracies": 0.574999988079071,
+      "rewards/generated": 0.7121507525444031,
+      "rewards/margins": 0.17485690116882324,
+      "rewards/real": 0.8870077133178711,
+      "step": 20
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 93.90039899223318,
+      "learning_rate": 2.909090909090909e-07,
+      "logits/generated": -2.5432677268981934,
+      "logits/real": -2.6215310096740723,
+      "logps/generated": -104.48051452636719,
+      "logps/real": -110.7447280883789,
+      "loss": 0.7882,
+      "rewards/accuracies": 0.6625000238418579,
+      "rewards/generated": 0.9083749651908875,
+      "rewards/margins": 0.3334098756313324,
+      "rewards/real": 1.2417848110198975,
+      "step": 30
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 78.65049485064348,
+      "learning_rate": 2e-07,
+      "logits/generated": -2.698758602142334,
+      "logits/real": -2.674772024154663,
+      "logps/generated": -115.9805679321289,
+      "logps/real": -127.0026626586914,
+      "loss": 0.8294,
+      "rewards/accuracies": 0.5625,
+      "rewards/generated": 1.7706496715545654,
+      "rewards/margins": 0.20217832922935486,
+      "rewards/real": 1.9728281497955322,
+      "step": 40
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 89.7604278601863,
+      "learning_rate": 1.0909090909090908e-07,
+      "logits/generated": -2.7130448818206787,
+      "logits/real": -2.705726146697998,
+      "logps/generated": -114.24468994140625,
+      "logps/real": -130.27967834472656,
+      "loss": 0.8066,
+      "rewards/accuracies": 0.5874999761581421,
+      "rewards/generated": 1.8750594854354858,
+      "rewards/margins": 0.24802803993225098,
+      "rewards/real": 2.1230876445770264,
+      "step": 50
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 83.07403448869721,
+      "learning_rate": 1.818181818181818e-08,
+      "logits/generated": -2.614271879196167,
+      "logits/real": -2.6497485637664795,
+      "logps/generated": -119.13523864746094,
+      "logps/real": -124.77104187011719,
+      "loss": 0.7921,
+      "rewards/accuracies": 0.625,
+      "rewards/generated": 1.9528534412384033,
+      "rewards/margins": 0.3087817132472992,
+      "rewards/real": 2.2616350650787354,
+      "step": 60
+    },
+    {
+      "epoch": 0.992,
+      "step": 62,
+      "total_flos": 0.0,
+      "train_loss": 0.8068127228367713,
+      "train_runtime": 741.4049,
+      "train_samples_per_second": 2.698,
+      "train_steps_per_second": 0.084
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 62,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}