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-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-v0.1-spin-2k-hhrlhf
+
+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": 0.11703237614804698,
+    "train_runtime": 741.0872,
+    "train_samples": 1999,
+    "train_samples_per_second": 2.697,
+    "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": 0.11703237614804698,
+    "train_runtime": 741.0872,
+    "train_samples": 1999,
+    "train_samples_per_second": 2.697,
+    "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": 368.63389421513955,
+      "learning_rate": 7.142857142857142e-08,
+      "logits/generated": -2.9934349060058594,
+      "logits/real": -2.747741222381592,
+      "logps/generated": -240.1473388671875,
+      "logps/real": -93.4220199584961,
+      "loss": 0.7913,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 1.784860986873838,
+      "learning_rate": 4.727272727272727e-07,
+      "logits/generated": -2.8636107444763184,
+      "logits/real": -2.1776232719421387,
+      "logps/generated": -275.1824951171875,
+      "logps/real": -95.4654769897461,
+      "loss": 0.2707,
+      "rewards/accuracies": 0.8888888955116272,
+      "rewards/generated": -3.9822473526000977,
+      "rewards/margins": 5.366815090179443,
+      "rewards/real": 1.3845678567886353,
+      "step": 10
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 1.195516804976816,
+      "learning_rate": 3.818181818181818e-07,
+      "logits/generated": -3.074063777923584,
+      "logits/real": -2.132559299468994,
+      "logps/generated": -325.47174072265625,
+      "logps/real": -77.88875579833984,
+      "loss": 0.0846,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -9.100537300109863,
+      "rewards/margins": 12.047636985778809,
+      "rewards/real": 2.947101593017578,
+      "step": 20
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.781989470530369,
+      "learning_rate": 2.909090909090909e-07,
+      "logits/generated": -3.0295379161834717,
+      "logits/real": -2.112544536590576,
+      "logps/generated": -355.9772033691406,
+      "logps/real": -76.08865356445312,
+      "loss": 0.0782,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.466151237487793,
+      "rewards/margins": 14.567548751831055,
+      "rewards/real": 3.1013970375061035,
+      "step": 30
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.9655945444920763,
+      "learning_rate": 2e-07,
+      "logits/generated": -3.012608051300049,
+      "logits/real": -2.2118849754333496,
+      "logps/generated": -348.3521423339844,
+      "logps/real": -83.44706726074219,
+      "loss": 0.078,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.384729385375977,
+      "rewards/margins": 14.619766235351562,
+      "rewards/real": 3.235036849975586,
+      "step": 40
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1.4386865003268576,
+      "learning_rate": 1.0909090909090908e-07,
+      "logits/generated": -3.0476737022399902,
+      "logits/real": -2.1677966117858887,
+      "logps/generated": -343.5052185058594,
+      "logps/real": -82.69391632080078,
+      "loss": 0.0699,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.07672119140625,
+      "rewards/margins": 14.31847095489502,
+      "rewards/real": 3.2417500019073486,
+      "step": 50
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.917235829369125,
+      "learning_rate": 1.818181818181818e-08,
+      "logits/generated": -3.0156235694885254,
+      "logits/real": -2.1144680976867676,
+      "logps/generated": -349.30865478515625,
+      "logps/real": -80.76779174804688,
+      "loss": 0.0799,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.362370491027832,
+      "rewards/margins": 14.571017265319824,
+      "rewards/real": 3.208648681640625,
+      "step": 60
+    },
+    {
+      "epoch": 0.992,
+      "step": 62,
+      "total_flos": 0.0,
+      "train_loss": 0.11703237614804698,
+      "train_runtime": 741.0872,
+      "train_samples_per_second": 2.697,
+      "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
+}