Model save

README.md

@@ -0,0 +1,58 @@
+---
+license: llama2
+base_model: meta-llama/Llama-2-7b-hf
+tags:
+- generated_from_trainer
+model-index:
+- name: Llama-2-7b-gen-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. -->
+
+# Llama-2-7b-gen-dpo-2k
+
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) 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": 0.8720041565273119,
+    "train_runtime": 584.014,
+    "train_samples": 1466,
+    "train_samples_per_second": 2.51,
+    "train_steps_per_second": 0.079
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.43.3"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 0.0,
+    "train_loss": 0.8720041565273119,
+    "train_runtime": 584.014,
+    "train_samples": 1466,
+    "train_samples_per_second": 2.51,
+    "train_steps_per_second": 0.079
+}

trainer_state.json

@@ -0,0 +1,117 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 200,
+  "global_step": 46,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021739130434782608,
+      "grad_norm": 49.405524700479546,
+      "learning_rate": 1e-07,
+      "logits/generated": -0.6996344327926636,
+      "logits/real": -0.6411839723587036,
+      "logps/generated": -268.8082580566406,
+      "logps/real": -278.0677490234375,
+      "loss": 0.9671,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.21739130434782608,
+      "grad_norm": 41.54584385387338,
+      "learning_rate": 4.390243902439024e-07,
+      "logits/generated": -0.8907153010368347,
+      "logits/real": -0.8477872014045715,
+      "logps/generated": -284.6246337890625,
+      "logps/real": -273.05023193359375,
+      "loss": 0.972,
+      "rewards/accuracies": 0.4444444477558136,
+      "rewards/generated": 0.16897444427013397,
+      "rewards/margins": -0.029052892699837685,
+      "rewards/real": 0.13992153108119965,
+      "step": 10
+    },
+    {
+      "epoch": 0.43478260869565216,
+      "grad_norm": 34.40402976139953,
+      "learning_rate": 3.170731707317073e-07,
+      "logits/generated": -0.8808904886245728,
+      "logits/real": -0.8196160197257996,
+      "logps/generated": -291.40716552734375,
+      "logps/real": -271.150634765625,
+      "loss": 0.8645,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/generated": 0.16139307618141174,
+      "rewards/margins": 0.2521939277648926,
+      "rewards/real": 0.4135870039463043,
+      "step": 20
+    },
+    {
+      "epoch": 0.6521739130434783,
+      "grad_norm": 33.980993784539564,
+      "learning_rate": 1.951219512195122e-07,
+      "logits/generated": -0.8029176592826843,
+      "logits/real": -0.7914024591445923,
+      "logps/generated": -294.6733093261719,
+      "logps/real": -278.3045349121094,
+      "loss": 0.8687,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/generated": 0.16676124930381775,
+      "rewards/margins": 0.27672332525253296,
+      "rewards/real": 0.4434846043586731,
+      "step": 30
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 51.08625886158765,
+      "learning_rate": 7.317073170731706e-08,
+      "logits/generated": -0.8437131643295288,
+      "logits/real": -0.8044729232788086,
+      "logps/generated": -295.627685546875,
+      "logps/real": -277.035400390625,
+      "loss": 0.8454,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/generated": 0.19026382267475128,
+      "rewards/margins": 0.39240655303001404,
+      "rewards/real": 0.5826703906059265,
+      "step": 40
+    },
+    {
+      "epoch": 1.0,
+      "step": 46,
+      "total_flos": 0.0,
+      "train_loss": 0.8720041565273119,
+      "train_runtime": 584.014,
+      "train_samples_per_second": 2.51,
+      "train_steps_per_second": 0.079
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 46,
+  "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
+}