Model save

README.md

@@ -0,0 +1,58 @@
+---
+license: apache-2.0
+base_model: mistralai/Mistral-7B-v0.3
+tags:
+- generated_from_trainer
+model-index:
+- name: Mistral-7B-v0.3-spin-2k-hhrlhf
+  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-v0.3-spin-2k-hhrlhf
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) 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": 0.11663837057928886,
+    "train_runtime": 749.8498,
+    "train_samples": 1999,
+    "train_samples_per_second": 2.666,
+    "train_steps_per_second": 0.083
+}

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": 0.11663837057928886,
+    "train_runtime": 749.8498,
+    "train_samples": 1999,
+    "train_samples_per_second": 2.666,
+    "train_steps_per_second": 0.083
+}

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": 270.6478620436586,
+      "learning_rate": 7.142857142857142e-08,
+      "logits/generated": -2.8518388271331787,
+      "logits/real": -2.431556224822998,
+      "logps/generated": -241.12493896484375,
+      "logps/real": -88.04447174072266,
+      "loss": 0.788,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 3.4825336012871664,
+      "learning_rate": 4.727272727272727e-07,
+      "logits/generated": -3.112872362136841,
+      "logits/real": -2.3448193073272705,
+      "logps/generated": -285.2105712890625,
+      "logps/real": -104.45608520507812,
+      "loss": 0.279,
+      "rewards/accuracies": 0.8888888955116272,
+      "rewards/generated": -3.95505690574646,
+      "rewards/margins": 5.338830471038818,
+      "rewards/real": 1.3837733268737793,
+      "step": 10
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.900671571621766,
+      "learning_rate": 3.818181818181818e-07,
+      "logits/generated": -3.139878749847412,
+      "logits/real": -2.3643927574157715,
+      "logps/generated": -320.77362060546875,
+      "logps/real": -73.03101348876953,
+      "loss": 0.0785,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -8.526379585266113,
+      "rewards/margins": 11.384855270385742,
+      "rewards/real": 2.858475923538208,
+      "step": 20
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.4149577906908892,
+      "learning_rate": 2.909090909090909e-07,
+      "logits/generated": -3.261312484741211,
+      "logits/real": -2.3303236961364746,
+      "logps/generated": -352.6148376464844,
+      "logps/real": -65.670654296875,
+      "loss": 0.0706,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -10.558695793151855,
+      "rewards/margins": 13.603212356567383,
+      "rewards/real": 3.0445168018341064,
+      "step": 30
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 1.78406972696896,
+      "learning_rate": 2e-07,
+      "logits/generated": -3.309864044189453,
+      "logits/real": -2.4957680702209473,
+      "logps/generated": -344.9488525390625,
+      "logps/real": -93.11787414550781,
+      "loss": 0.0814,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -10.671560287475586,
+      "rewards/margins": 14.303810119628906,
+      "rewards/real": 3.632251262664795,
+      "step": 40
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.918331726023657,
+      "learning_rate": 1.0909090909090908e-07,
+      "logits/generated": -3.2701480388641357,
+      "logits/real": -2.330775737762451,
+      "logps/generated": -352.88299560546875,
+      "logps/real": -85.17743682861328,
+      "loss": 0.0762,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -10.926799774169922,
+      "rewards/margins": 14.398449897766113,
+      "rewards/real": 3.4716498851776123,
+      "step": 50
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.8334083410423753,
+      "learning_rate": 1.818181818181818e-08,
+      "logits/generated": -3.3187618255615234,
+      "logits/real": -2.187263011932373,
+      "logps/generated": -343.95556640625,
+      "logps/real": -66.73863983154297,
+      "loss": 0.0748,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -10.801365852355957,
+      "rewards/margins": 14.124551773071289,
+      "rewards/real": 3.3231849670410156,
+      "step": 60
+    },
+    {
+      "epoch": 0.992,
+      "step": 62,
+      "total_flos": 0.0,
+      "train_loss": 0.11663837057928886,
+      "train_runtime": 749.8498,
+      "train_samples_per_second": 2.666,
+      "train_steps_per_second": 0.083
+    }
+  ],
+  "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
+}