Training in progress, step 2072, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:857959d85d07261444eb5e7afcb45a95c7b787da4c54ceb28ccef0d2b4133568
+oid sha256:3b3348347c5794cd8071f0cef1cc9e1495550a9f8017f064f8d08001da935de6
 size 232169792

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f418ad7f0a24001cc34d54f0e84561b2198f66e0016cf458ada0bdc754ede411
+oid sha256:6294f772f9da965b0dc3c23ee7713331071f97aea88c9f208bf5e6c9239b1585
 size 117446154

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2ecf95f7fbf9c676ce35814bbd401c8f1c386fc419d0d2d22bb241006b77c48
+oid sha256:35f511fc6ceb095d52923c4d373851bf545d0c675a13d2bfda0c9c62cf0ead84
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.9922929144536106,
+  "epoch": 2.998078878969375,
   "eval_steps": 500,
-  "global_step": 2068,
+  "global_step": 2072,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -37231,6 +37231,78 @@
       "rewards/margins": 0.19329190254211426,
       "rewards/rejected": -0.28255730867385864,
       "step": 2068
+    },
+    {
+      "epoch": 2.9937394055825517,
+      "grad_norm": 0.5331955552101135,
+      "learning_rate": 4.5933528375341485e-10,
+      "log_odds_chosen": 2.6847963333129883,
+      "log_odds_ratio": -0.4565506875514984,
+      "logits/chosen": -1.810483694076538,
+      "logits/rejected": -1.6029425859451294,
+      "logps/chosen": -0.806846022605896,
+      "logps/rejected": -2.9770960807800293,
+      "loss": 1.06,
+      "nll_loss": 1.0143355131149292,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.0806846022605896,
+      "rewards/margins": 0.2170250117778778,
+      "rewards/rejected": -0.2977096438407898,
+      "step": 2069
+    },
+    {
+      "epoch": 2.995185896711493,
+      "grad_norm": 0.5308060050010681,
+      "learning_rate": 2.58376443265651e-10,
+      "log_odds_chosen": 2.103809118270874,
+      "log_odds_ratio": -0.5330842733383179,
+      "logits/chosen": -1.7134472131729126,
+      "logits/rejected": -1.536543846130371,
+      "logps/chosen": -0.7203590869903564,
+      "logps/rejected": -2.3935389518737793,
+      "loss": 1.015,
+      "nll_loss": 0.9616963863372803,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.07203590869903564,
+      "rewards/margins": 0.16731798648834229,
+      "rewards/rejected": -0.23935389518737793,
+      "step": 2070
+    },
+    {
+      "epoch": 2.996632387840434,
+      "grad_norm": 1.1535701751708984,
+      "learning_rate": 1.1483408467460876e-10,
+      "log_odds_chosen": 2.2093892097473145,
+      "log_odds_ratio": -0.5061578154563904,
+      "logits/chosen": -1.8256726264953613,
+      "logits/rejected": -1.6192009449005127,
+      "logps/chosen": -0.8687729239463806,
+      "logps/rejected": -2.67683482170105,
+      "loss": 1.1116,
+      "nll_loss": 1.0610023736953735,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.08687728643417358,
+      "rewards/margins": 0.18080618977546692,
+      "rewards/rejected": -0.2676834762096405,
+      "step": 2071
+    },
+    {
+      "epoch": 2.998078878969375,
+      "grad_norm": 0.5611602663993835,
+      "learning_rate": 2.8708537652688548e-11,
+      "log_odds_chosen": 1.6254621744155884,
+      "log_odds_ratio": -0.5093374252319336,
+      "logits/chosen": -1.76937997341156,
+      "logits/rejected": -1.670184850692749,
+      "logps/chosen": -0.8887354135513306,
+      "logps/rejected": -2.101914882659912,
+      "loss": 1.1095,
+      "nll_loss": 1.0585343837738037,
+      "rewards/accuracies": 0.734375,
+      "rewards/chosen": -0.08887353539466858,
+      "rewards/margins": 0.12131794542074203,
+      "rewards/rejected": -0.2101914882659912,
+      "step": 2072
     }
   ],
   "logging_steps": 1,