Training in progress, step 1980, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72de3d2cd81e75f4ecc70f5acd430ff5b3122913c0c27640b64e85b0a53935be
+oid sha256:c55df82de4fb0d3db68714ad442d1d38d6669acfa32b0ffab80c2d7a3ba49721
 size 232169792

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9e31b34b8d1af3114dd3b2085b21005ecf4d64b9404a12f78a6f1bda698b0b3
+oid sha256:e0132de8f62e0022bee7ed5782fad90b3e6cee029fe97753323c522d57dad933
 size 117446154

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2b0fa43290a459cc7417471e7cdda585e9a0e5edf2c2c7ef8247d02c7ef7b67
+oid sha256:d4df863dffbf5bc9bd29858a1d6f78b8ba5687345e29875e39b5ed6c18141865
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.859215730591027,
+  "epoch": 2.8650016951067916,
   "eval_steps": 500,
-  "global_step": 1976,
+  "global_step": 1980,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -35575,6 +35575,78 @@
       "rewards/margins": 0.32102257013320923,
       "rewards/rejected": -0.3932061493396759,
       "step": 1976
+    },
+    {
+      "epoch": 2.860662221719968,
+      "grad_norm": 0.5800076127052307,
+      "learning_rate": 2.641115862252713e-07,
+      "log_odds_chosen": 3.0793051719665527,
+      "log_odds_ratio": -0.4249449074268341,
+      "logits/chosen": -1.7792483568191528,
+      "logits/rejected": -1.466111183166504,
+      "logps/chosen": -0.7235685586929321,
+      "logps/rejected": -3.3135576248168945,
+      "loss": 0.9257,
+      "nll_loss": 0.8831884860992432,
+      "rewards/accuracies": 0.765625,
+      "rewards/chosen": -0.07235686480998993,
+      "rewards/margins": 0.25899893045425415,
+      "rewards/rejected": -0.3313557803630829,
+      "step": 1977
+    },
+    {
+      "epoch": 2.8621087128489093,
+      "grad_norm": 0.6505364179611206,
+      "learning_rate": 2.586473775788856e-07,
+      "log_odds_chosen": 1.9093005657196045,
+      "log_odds_ratio": -0.4956282377243042,
+      "logits/chosen": -1.8862462043762207,
+      "logits/rejected": -1.6489293575286865,
+      "logps/chosen": -0.8198176622390747,
+      "logps/rejected": -2.364813804626465,
+      "loss": 1.0495,
+      "nll_loss": 0.9999848008155823,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.08198177814483643,
+      "rewards/margins": 0.1544996052980423,
+      "rewards/rejected": -0.23648138344287872,
+      "step": 1978
+    },
+    {
+      "epoch": 2.8635552039778505,
+      "grad_norm": 1.2835837602615356,
+      "learning_rate": 2.5323999197676973e-07,
+      "log_odds_chosen": 1.9394054412841797,
+      "log_odds_ratio": -0.4927099645137787,
+      "logits/chosen": -1.8224869966506958,
+      "logits/rejected": -1.6263892650604248,
+      "logps/chosen": -0.8234947919845581,
+      "logps/rejected": -2.4225282669067383,
+      "loss": 1.0989,
+      "nll_loss": 1.0496528148651123,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.08234947919845581,
+      "rewards/margins": 0.15990334749221802,
+      "rewards/rejected": -0.24225284159183502,
+      "step": 1979
+    },
+    {
+      "epoch": 2.8650016951067916,
+      "grad_norm": 0.6098787784576416,
+      "learning_rate": 2.478894418379674e-07,
+      "log_odds_chosen": 3.68991756439209,
+      "log_odds_ratio": -0.4527060389518738,
+      "logits/chosen": -1.8065725564956665,
+      "logits/rejected": -1.5102980136871338,
+      "logps/chosen": -0.761375367641449,
+      "logps/rejected": -3.9927573204040527,
+      "loss": 0.9802,
+      "nll_loss": 0.9349774122238159,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.07613754272460938,
+      "rewards/margins": 0.32313817739486694,
+      "rewards/rejected": -0.3992757499217987,
+      "step": 1980
     }
   ],
   "logging_steps": 1,