Model save

README.md

@@ -0,0 +1,58 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen2.5-7B
+tags:
+- generated_from_trainer
+model-index:
+- name: Qwen2.5-7B-gen-dpo-10k
+  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. -->
+
+# Qwen2.5-7B-gen-dpo-10k
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) 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": 1.0,
+    "total_flos": 0.0,
+    "train_loss": 0.7470545699963203,
+    "train_runtime": 930.2413,
+    "train_samples": 2484,
+    "train_samples_per_second": 2.67,
+    "train_steps_per_second": 0.084
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.43.3"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 0.0,
+    "train_loss": 0.7470545699963203,
+    "train_runtime": 930.2413,
+    "train_samples": 2484,
+    "train_samples_per_second": 2.67,
+    "train_steps_per_second": 0.084
+}

trainer_state.json

@@ -0,0 +1,162 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 200,
+  "global_step": 78,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01282051282051282,
+      "grad_norm": 101.89968076534468,
+      "learning_rate": 6.25e-08,
+      "logits/generated": -1.0679720640182495,
+      "logits/real": -0.826486349105835,
+      "logps/generated": -273.1225891113281,
+      "logps/real": -293.10797119140625,
+      "loss": 0.9699,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.1282051282051282,
+      "grad_norm": 143.51099887453523,
+      "learning_rate": 4.857142857142857e-07,
+      "logits/generated": -0.8684251308441162,
+      "logits/real": -0.6457947492599487,
+      "logps/generated": -273.63232421875,
+      "logps/real": -274.0804748535156,
+      "loss": 0.9133,
+      "rewards/accuracies": 0.625,
+      "rewards/generated": 0.16648532450199127,
+      "rewards/margins": 0.10824514925479889,
+      "rewards/real": 0.27473047375679016,
+      "step": 10
+    },
+    {
+      "epoch": 0.2564102564102564,
+      "grad_norm": 61.070639184089565,
+      "learning_rate": 4.142857142857143e-07,
+      "logits/generated": -1.0700271129608154,
+      "logits/real": -0.7057520747184753,
+      "logps/generated": -261.4720153808594,
+      "logps/real": -257.3920593261719,
+      "loss": 0.7717,
+      "rewards/accuracies": 0.75,
+      "rewards/generated": 0.8906627893447876,
+      "rewards/margins": 0.7612454295158386,
+      "rewards/real": 1.6519081592559814,
+      "step": 20
+    },
+    {
+      "epoch": 0.38461538461538464,
+      "grad_norm": 64.32897890365741,
+      "learning_rate": 3.4285714285714286e-07,
+      "logits/generated": -0.9067907333374023,
+      "logits/real": -0.659667432308197,
+      "logps/generated": -261.84893798828125,
+      "logps/real": -250.3210906982422,
+      "loss": 0.7144,
+      "rewards/accuracies": 0.75,
+      "rewards/generated": 1.7948137521743774,
+      "rewards/margins": 0.8462657928466797,
+      "rewards/real": 2.6410794258117676,
+      "step": 30
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 164.54242317315587,
+      "learning_rate": 2.714285714285714e-07,
+      "logits/generated": -0.9878751039505005,
+      "logits/real": -0.6697134375572205,
+      "logps/generated": -254.94430541992188,
+      "logps/real": -251.97024536132812,
+      "loss": 0.7095,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/generated": 1.6531692743301392,
+      "rewards/margins": 1.1178901195526123,
+      "rewards/real": 2.771059513092041,
+      "step": 40
+    },
+    {
+      "epoch": 0.6410256410256411,
+      "grad_norm": 65.89825293961819,
+      "learning_rate": 2e-07,
+      "logits/generated": -0.9043434262275696,
+      "logits/real": -0.767203152179718,
+      "logps/generated": -254.6318817138672,
+      "logps/real": -240.71481323242188,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/generated": 2.046715497970581,
+      "rewards/margins": 0.7744835615158081,
+      "rewards/real": 2.8211989402770996,
+      "step": 50
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 124.30731613732516,
+      "learning_rate": 1.2857142857142855e-07,
+      "logits/generated": -1.0582646131515503,
+      "logits/real": -0.6999293565750122,
+      "logps/generated": -245.7989044189453,
+      "logps/real": -248.5264129638672,
+      "loss": 0.7132,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/generated": 1.8123178482055664,
+      "rewards/margins": 1.1895701885223389,
+      "rewards/real": 3.0018882751464844,
+      "step": 60
+    },
+    {
+      "epoch": 0.8974358974358975,
+      "grad_norm": 65.42190072046155,
+      "learning_rate": 5.714285714285714e-08,
+      "logits/generated": -1.1064507961273193,
+      "logits/real": -0.7838689684867859,
+      "logps/generated": -245.507568359375,
+      "logps/real": -237.945068359375,
+      "loss": 0.7122,
+      "rewards/accuracies": 0.75,
+      "rewards/generated": 1.7316808700561523,
+      "rewards/margins": 1.1938471794128418,
+      "rewards/real": 2.925528049468994,
+      "step": 70
+    },
+    {
+      "epoch": 1.0,
+      "step": 78,
+      "total_flos": 0.0,
+      "train_loss": 0.7470545699963203,
+      "train_runtime": 930.2413,
+      "train_samples_per_second": 2.67,
+      "train_steps_per_second": 0.084
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 78,
+  "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
+}