Training in progress, epoch 0

adapter_config.json

@@ -16,10 +16,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "q_proj",
     "v_proj",
-    "k_proj"
+    "q_proj",
+    "k_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fcd987e4295cf39204d499f3d64718c79f91cd5edcd56838915d9b76fbfe4dc
+oid sha256:b406044886581aee396c928a6e0154eb09cc26550ff19e5930abfb44a682b43b
 size 218138576

all_results.json

@@ -1,21 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -3.0078587532043457,
-    "eval_logits/rejected": -3.042999744415283,
-    "eval_logps/chosen": -194.36697387695312,
-    "eval_logps/rejected": -179.82501220703125,
-    "eval_loss": 0.6931471824645996,
-    "eval_rewards/accuracies": 0.0,
-    "eval_rewards/chosen": 0.0,
-    "eval_rewards/margins": 0.0,
-    "eval_rewards/rejected": 0.0,
-    "eval_runtime": 5.0964,
-    "eval_samples": 10,
-    "eval_samples_per_second": 1.962,
-    "eval_steps_per_second": 0.392,
-    "train_loss": 0.3465735912322998,
-    "train_runtime": 16.8017,
+    "train_loss": 19.057324022054672,
+    "train_runtime": 626.0842,
     "train_samples": 61966,
-    "train_samples_per_second": 0.595,
-    "train_steps_per_second": 0.06
+    "train_samples_per_second": 0.799,
+    "train_steps_per_second": 0.026
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "train_loss": 0.3465735912322998,
-    "train_runtime": 16.8017,
+    "train_loss": 19.057324022054672,
+    "train_runtime": 626.0842,
     "train_samples": 61966,
-    "train_samples_per_second": 0.595,
-    "train_steps_per_second": 0.06
+    "train_samples_per_second": 0.799,
+    "train_steps_per_second": 0.026
 }

trainer_state.json

@@ -3,53 +3,67 @@
   "best_model_checkpoint": null,
   "epoch": 1.0,
   "eval_steps": 100,
-  "global_step": 1,
+  "global_step": 16,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "learning_rate": 0.0,
-      "logits/chosen": -3.053490400314331,
-      "logits/rejected": -3.094686985015869,
-      "logps/chosen": -181.32530212402344,
-      "logps/rejected": -175.11618041992188,
-      "loss": 0.3466,
+      "epoch": 0.06,
+      "learning_rate": 0.00025,
+      "logits/chosen": -3.027029514312744,
+      "logits/rejected": -3.115102767944336,
+      "logps/chosen": -313.0743103027344,
+      "logps/rejected": -258.258544921875,
+      "loss": 0.6931,
       "rewards/accuracies": 0.0,
       "rewards/chosen": 0.0,
       "rewards/margins": 0.0,
       "rewards/rejected": 0.0,
       "step": 1
     },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00021428571428571427,
+      "logits/chosen": -2.8371262550354004,
+      "logits/rejected": -2.8233795166015625,
+      "logps/chosen": -797.5036010742188,
+      "logps/rejected": -571.7381591796875,
+      "loss": 25.7534,
+      "rewards/accuracies": 0.3819444477558136,
+      "rewards/chosen": -54.44696807861328,
+      "rewards/margins": -20.015228271484375,
+      "rewards/rejected": -34.43173599243164,
+      "step": 10
+    },
     {
       "epoch": 1.0,
-      "eval_logits/chosen": -3.0078587532043457,
-      "eval_logits/rejected": -3.042999744415283,
-      "eval_logps/chosen": -194.36697387695312,
-      "eval_logps/rejected": -179.82501220703125,
-      "eval_loss": 0.6931471824645996,
-      "eval_rewards/accuracies": 0.0,
-      "eval_rewards/chosen": 0.0,
-      "eval_rewards/margins": 0.0,
-      "eval_rewards/rejected": 0.0,
-      "eval_runtime": 5.1666,
-      "eval_samples_per_second": 1.936,
-      "eval_steps_per_second": 0.387,
-      "step": 1
+      "eval_logits/chosen": -3.1348018646240234,
+      "eval_logits/rejected": -3.0953807830810547,
+      "eval_logps/chosen": -279.9947509765625,
+      "eval_logps/rejected": -279.706787109375,
+      "eval_loss": 0.947689950466156,
+      "eval_rewards/accuracies": 0.6944444179534912,
+      "eval_rewards/chosen": -3.136577844619751,
+      "eval_rewards/margins": 2.351809501647949,
+      "eval_rewards/rejected": -5.488387107849121,
+      "eval_runtime": 198.8798,
+      "eval_samples_per_second": 2.514,
+      "eval_steps_per_second": 0.317,
+      "step": 16
     },
     {
       "epoch": 1.0,
-      "step": 1,
+      "step": 16,
       "total_flos": 0.0,
-      "train_loss": 0.3465735912322998,
-      "train_runtime": 16.8017,
-      "train_samples_per_second": 0.595,
-      "train_steps_per_second": 0.06
+      "train_loss": 19.057324022054672,
+      "train_runtime": 626.0842,
+      "train_samples_per_second": 0.799,
+      "train_steps_per_second": 0.026
     }
   ],
   "logging_steps": 10,
-  "max_steps": 1,
+  "max_steps": 16,
   "num_train_epochs": 1,
   "save_steps": 500,
   "total_flos": 0.0,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1822fa9c0a896388f17c4a6fafc12ec68e5d9a0b52d64a949b92685ec29039db
+oid sha256:70f9924616298b428922bcac8ef53f857cc390c4cc29b39f72d9ef1110712bfd
 size 4792