Training in progress, step 8, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b6511c5cdd67e13fe656885b0de5ee8e6bb936e219652f4559ed66017a7f447
+oid sha256:3b1de24cbfa9f4cd9b6b61603c5c430d66f5117b0066916e6a94348617ac0ab3
 size 83945296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e69e9eb99bb2e67bbd5c59acbbd058da710ee35a726b98f2acc46fcaa276f269
+oid sha256:0b68b8c960390833091b8923f7bf648b58f1b9f278a335e629b104d215fc140a
 size 42545748

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7adb031dc938779259a07e684ea318126e92e52b40c256b93ef474d2ac57a5b
+oid sha256:6a3d18e607a27cfab9cec7cfaa7384cdb877a2330c4bcd4e1efcae25be9908cb
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.081554635234151,
+  "epoch": 0.163109270468302,
   "eval_steps": 500,
-  "global_step": 4,
+  "global_step": 8,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -79,6 +79,78 @@
       "rewards/margins": -0.006830527447164059,
       "rewards/rejected": -0.48478949069976807,
       "step": 4
+    },
+    {
+      "epoch": 0.10194329404268876,
+      "grad_norm": 9.331730842590332,
+      "learning_rate": 5e-05,
+      "log_odds_chosen": -0.07571306079626083,
+      "log_odds_ratio": -0.7841904163360596,
+      "logits/chosen": -2.3229925632476807,
+      "logits/rejected": -2.337273120880127,
+      "logps/chosen": -4.687668323516846,
+      "logps/rejected": -4.612335205078125,
+      "loss": 4.9652,
+      "nll_loss": 4.886806011199951,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.46876680850982666,
+      "rewards/margins": -0.0075332350097596645,
+      "rewards/rejected": -0.46123358607292175,
+      "step": 5
+    },
+    {
+      "epoch": 0.1223319528512265,
+      "grad_norm": 9.857681274414062,
+      "learning_rate": 6e-05,
+      "log_odds_chosen": -0.19471649825572968,
+      "log_odds_ratio": -0.8327277302742004,
+      "logits/chosen": -2.3558108806610107,
+      "logits/rejected": -2.3881752490997314,
+      "logps/chosen": -4.344555377960205,
+      "logps/rejected": -4.152055263519287,
+      "loss": 4.6379,
+      "nll_loss": 4.554634094238281,
+      "rewards/accuracies": 0.40625,
+      "rewards/chosen": -0.4344555735588074,
+      "rewards/margins": -0.019250018522143364,
+      "rewards/rejected": -0.41520553827285767,
+      "step": 6
+    },
+    {
+      "epoch": 0.14272061165976427,
+      "grad_norm": 7.811158657073975,
+      "learning_rate": 7e-05,
+      "log_odds_chosen": -0.0166391022503376,
+      "log_odds_ratio": -0.7380757331848145,
+      "logits/chosen": -2.388141632080078,
+      "logits/rejected": -2.4195711612701416,
+      "logps/chosen": -3.969034194946289,
+      "logps/rejected": -3.951852798461914,
+      "loss": 4.2265,
+      "nll_loss": 4.152710914611816,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.39690345525741577,
+      "rewards/margins": -0.0017181318253278732,
+      "rewards/rejected": -0.39518529176712036,
+      "step": 7
+    },
+    {
+      "epoch": 0.163109270468302,
+      "grad_norm": 7.66775369644165,
+      "learning_rate": 8e-05,
+      "log_odds_chosen": 0.02697194740176201,
+      "log_odds_ratio": -0.7153270840644836,
+      "logits/chosen": -2.477844476699829,
+      "logits/rejected": -2.474121570587158,
+      "logps/chosen": -3.4541428089141846,
+      "logps/rejected": -3.4769277572631836,
+      "loss": 3.6939,
+      "nll_loss": 3.6224091053009033,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.34541425108909607,
+      "rewards/margins": 0.0022785186301916838,
+      "rewards/rejected": -0.3476927876472473,
+      "step": 8
     }
   ],
   "logging_steps": 1,