Training in progress, step 36, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3163c12a6c0639e3c623879492b912a343b23c6b92d7995dda7af17679dfd574
+oid sha256:32a880d878e6aa79b0300377e717478cfa772b82f2e914e8a2da3c7d55b9b67b
 size 590925768

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4aa88cded1b216ede4cdf2de0acc168aa0ceed1fb2880577eee7d9d834acda77
+oid sha256:70af22ed26a3400885dc57c6f2ee5f96dab909884a0762ad968e4ea440cfa3bd
 size 296481140

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f208e74f7004e2ce764e7d449d5f545644604ac97c4cc7a8df5e4e4ca120e55c
+oid sha256:b5593ad2d1e2b9381aaf428be189e25ebe80f3dc86e44e105dfdab3b2565f9a3
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.04675478848480698,
+  "epoch": 0.05259913704540785,
   "eval_steps": 500,
-  "global_step": 32,
+  "global_step": 36,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -583,6 +583,78 @@
       "rewards/margins": 0.00609829043969512,
       "rewards/rejected": -0.12037733197212219,
       "step": 32
+    },
+    {
+      "epoch": 0.048215875624957195,
+      "grad_norm": 0.4958839416503906,
+      "learning_rate": 9.963651857945286e-06,
+      "log_odds_chosen": -0.07793000340461731,
+      "log_odds_ratio": -0.7607824206352234,
+      "logits/chosen": -2.067959785461426,
+      "logits/rejected": -2.06540584564209,
+      "logps/chosen": -1.3159579038619995,
+      "logps/rejected": -1.2743453979492188,
+      "loss": 1.657,
+      "nll_loss": 1.5809520483016968,
+      "rewards/accuracies": 0.390625,
+      "rewards/chosen": -0.13159577548503876,
+      "rewards/margins": -0.004161248914897442,
+      "rewards/rejected": -0.12743453681468964,
+      "step": 33
+    },
+    {
+      "epoch": 0.04967696276510741,
+      "grad_norm": 0.530023992061615,
+      "learning_rate": 9.960805806375854e-06,
+      "log_odds_chosen": 0.017400667071342468,
+      "log_odds_ratio": -0.7287170886993408,
+      "logits/chosen": -2.049931526184082,
+      "logits/rejected": -2.067579984664917,
+      "logps/chosen": -1.313307285308838,
+      "logps/rejected": -1.3060802221298218,
+      "loss": 1.5791,
+      "nll_loss": 1.506237268447876,
+      "rewards/accuracies": 0.390625,
+      "rewards/chosen": -0.1313307285308838,
+      "rewards/margins": -0.0007226967136375606,
+      "rewards/rejected": -0.13060802221298218,
+      "step": 34
+    },
+    {
+      "epoch": 0.05113804990525763,
+      "grad_norm": 0.5267317891120911,
+      "learning_rate": 9.957852929571799e-06,
+      "log_odds_chosen": -0.08755096048116684,
+      "log_odds_ratio": -0.7537316679954529,
+      "logits/chosen": -2.0170559883117676,
+      "logits/rejected": -2.0241918563842773,
+      "logps/chosen": -1.3282378911972046,
+      "logps/rejected": -1.267946481704712,
+      "loss": 1.5959,
+      "nll_loss": 1.5205148458480835,
+      "rewards/accuracies": 0.375,
+      "rewards/chosen": -0.13282378017902374,
+      "rewards/margins": -0.006029147654771805,
+      "rewards/rejected": -0.12679465115070343,
+      "step": 35
+    },
+    {
+      "epoch": 0.05259913704540785,
+      "grad_norm": 0.5598548054695129,
+      "learning_rate": 9.954793291119917e-06,
+      "log_odds_chosen": -0.07335270196199417,
+      "log_odds_ratio": -0.7724658846855164,
+      "logits/chosen": -2.0005574226379395,
+      "logits/rejected": -1.985095739364624,
+      "logps/chosen": -1.2228074073791504,
+      "logps/rejected": -1.1555684804916382,
+      "loss": 1.5164,
+      "nll_loss": 1.4391915798187256,
+      "rewards/accuracies": 0.359375,
+      "rewards/chosen": -0.12228074669837952,
+      "rewards/margins": -0.006723896134644747,
+      "rewards/rejected": -0.11555685847997665,
+      "step": 36
     }
   ],
   "logging_steps": 1.0,