Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70082a784b3a40b80eef4107594f9ad4aa081bd6d9f8c106958c2ea614d54277
+oid sha256:b23b63a1ba8da58de7be23d27a64b09e82a4a5a5184f6f73c8e716906a1444ec
 size 590925768

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a32692aa2cd37cdb83fa8f39f049e7c3266de86872bd53e275966d8eba7e480
+oid sha256:e6f426b7b1c120fd6698ec8e3ddf7b15450626dd3e1c5dc7e97ad1ef6c817211
 size 296481140

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85789d24a030dd0676934dfc35b17ca73a64f42fa104badf9fe0a48fd11a328f
+oid sha256:b022caf5394258d9701ac5c9c6bf11912e3fbbf6b0f72e2f43a2a5772c71d718
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.02337739424240349,
+  "epoch": 0.02922174280300436,
   "eval_steps": 500,
-  "global_step": 16,
+  "global_step": 20,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -295,6 +295,78 @@
       "rewards/margins": -0.0027017316315323114,
       "rewards/rejected": -0.12293566763401031,
       "step": 16
+    },
+    {
+      "epoch": 0.024838481382553707,
+      "grad_norm": 0.6587896943092346,
+      "learning_rate": 9.994617494508811e-06,
+      "log_odds_chosen": -0.010003458708524704,
+      "log_odds_ratio": -0.7430539727210999,
+      "logits/chosen": -2.200150966644287,
+      "logits/rejected": -2.1906399726867676,
+      "logps/chosen": -1.2153267860412598,
+      "logps/rejected": -1.1962745189666748,
+      "loss": 1.5641,
+      "nll_loss": 1.4898183345794678,
+      "rewards/accuracies": 0.390625,
+      "rewards/chosen": -0.12153266370296478,
+      "rewards/margins": -0.001905218348838389,
+      "rewards/rejected": -0.119627445936203,
+      "step": 17
+    },
+    {
+      "epoch": 0.026299568522703924,
+      "grad_norm": 0.5830309987068176,
+      "learning_rate": 9.993487413792276e-06,
+      "log_odds_chosen": -0.1340780407190323,
+      "log_odds_ratio": -0.7850849628448486,
+      "logits/chosen": -2.1663918495178223,
+      "logits/rejected": -2.145009756088257,
+      "logps/chosen": -1.2580608129501343,
+      "logps/rejected": -1.188957929611206,
+      "loss": 1.5591,
+      "nll_loss": 1.4806358814239502,
+      "rewards/accuracies": 0.34375,
+      "rewards/chosen": -0.1258060783147812,
+      "rewards/margins": -0.006910297088325024,
+      "rewards/rejected": -0.11889579892158508,
+      "step": 18
+    },
+    {
+      "epoch": 0.027760655662854142,
+      "grad_norm": 0.5536736249923706,
+      "learning_rate": 9.992249804080372e-06,
+      "log_odds_chosen": -0.021063022315502167,
+      "log_odds_ratio": -0.7351462244987488,
+      "logits/chosen": -2.1682121753692627,
+      "logits/rejected": -2.163313150405884,
+      "logps/chosen": -1.2965900897979736,
+      "logps/rejected": -1.2690778970718384,
+      "loss": 1.6294,
+      "nll_loss": 1.5559338331222534,
+      "rewards/accuracies": 0.421875,
+      "rewards/chosen": -0.1296590119600296,
+      "rewards/margins": -0.0027512230444699526,
+      "rewards/rejected": -0.12690778076648712,
+      "step": 19
+    },
+    {
+      "epoch": 0.02922174280300436,
+      "grad_norm": 0.5634390711784363,
+      "learning_rate": 9.990904692023604e-06,
+      "log_odds_chosen": 0.025263303890824318,
+      "log_odds_ratio": -0.7174615263938904,
+      "logits/chosen": -2.26141357421875,
+      "logits/rejected": -2.250270366668701,
+      "logps/chosen": -1.3806607723236084,
+      "logps/rejected": -1.3875806331634521,
+      "loss": 1.7203,
+      "nll_loss": 1.6485247611999512,
+      "rewards/accuracies": 0.375,
+      "rewards/chosen": -0.13806606829166412,
+      "rewards/margins": 0.0006919947918504477,
+      "rewards/rejected": -0.1387580782175064,
+      "step": 20
     }
   ],
   "logging_steps": 1.0,