Training in progress, step 1984, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "q_proj",
-    "o_proj",
-    "gate_proj",
-    "v_proj",
+    "down_proj",
     "k_proj",
-    "down_proj"
+    "v_proj",
+    "o_proj",
+    "q_proj",
+    "up_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c55df82de4fb0d3db68714ad442d1d38d6669acfa32b0ffab80c2d7a3ba49721
+oid sha256:0a8df42a2c05cead50bc256fc821227a1c685d9421a56e6c6c225f8a79d954a0
 size 232169792

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0132de8f62e0022bee7ed5782fad90b3e6cee029fe97753323c522d57dad933
+oid sha256:9fd201181b6924563f49d1bab056ca8efb3154ceec6ba4e87c992e8c957190e6
 size 117446154

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:040e778408d0cf72b1709cf08f1f9dc8d42f5365f15fd6b6a9157322711b37d2
+oid sha256:eaee5b57bceca0a9dd281cd9ac702d8c230d90e4a749a0f30fcacc9028ab8d66
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4df863dffbf5bc9bd29858a1d6f78b8ba5687345e29875e39b5ed6c18141865
+oid sha256:4bd095dae7ffbf005aa92a66a4077dd3d117f3f949f0ac377ea50e1a29fda449
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.8650016951067916,
+  "epoch": 2.870787659622556,
   "eval_steps": 500,
-  "global_step": 1980,
+  "global_step": 1984,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -35647,6 +35647,78 @@
       "rewards/margins": 0.32313817739486694,
       "rewards/rejected": -0.3992757499217987,
       "step": 1980
+    },
+    {
+      "epoch": 2.8664481862357327,
+      "grad_norm": 0.5579022169113159,
+      "learning_rate": 2.4259573945100144e-07,
+      "log_odds_chosen": 2.1421194076538086,
+      "log_odds_ratio": -0.5158656239509583,
+      "logits/chosen": -1.849534511566162,
+      "logits/rejected": -1.6556165218353271,
+      "logps/chosen": -0.8209269642829895,
+      "logps/rejected": -2.6318562030792236,
+      "loss": 1.0822,
+      "nll_loss": 1.0306111574172974,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.08209270238876343,
+      "rewards/margins": 0.18109291791915894,
+      "rewards/rejected": -0.26318562030792236,
+      "step": 1981
+    },
+    {
+      "epoch": 2.867894677364674,
+      "grad_norm": 0.5723074674606323,
+      "learning_rate": 2.3735889697382996e-07,
+      "log_odds_chosen": 3.1368532180786133,
+      "log_odds_ratio": -0.4154900312423706,
+      "logits/chosen": -1.7287541627883911,
+      "logits/rejected": -1.3899829387664795,
+      "logps/chosen": -0.7958812713623047,
+      "logps/rejected": -3.4790496826171875,
+      "loss": 0.9691,
+      "nll_loss": 0.9275829792022705,
+      "rewards/accuracies": 0.8125,
+      "rewards/chosen": -0.07958813011646271,
+      "rewards/margins": 0.2683168351650238,
+      "rewards/rejected": -0.3479049801826477,
+      "step": 1982
+    },
+    {
+      "epoch": 2.869341168493615,
+      "grad_norm": 0.5695856809616089,
+      "learning_rate": 2.3217892643381256e-07,
+      "log_odds_chosen": 1.77284574508667,
+      "log_odds_ratio": -0.4833868145942688,
+      "logits/chosen": -1.7728116512298584,
+      "logits/rejected": -1.57659113407135,
+      "logps/chosen": -0.7696139216423035,
+      "logps/rejected": -2.138911008834839,
+      "loss": 1.0594,
+      "nll_loss": 1.0110141038894653,
+      "rewards/accuracies": 0.703125,
+      "rewards/chosen": -0.07696139067411423,
+      "rewards/margins": 0.1369297057390213,
+      "rewards/rejected": -0.21389110386371613,
+      "step": 1983
+    },
+    {
+      "epoch": 2.870787659622556,
+      "grad_norm": 0.5508972406387329,
+      "learning_rate": 2.2705583972770783e-07,
+      "log_odds_chosen": 3.5796892642974854,
+      "log_odds_ratio": -0.39506861567497253,
+      "logits/chosen": -1.8140000104904175,
+      "logits/rejected": -1.4643973112106323,
+      "logps/chosen": -0.8209943771362305,
+      "logps/rejected": -3.885220527648926,
+      "loss": 1.0249,
+      "nll_loss": 0.9854146242141724,
+      "rewards/accuracies": 0.765625,
+      "rewards/chosen": -0.08209944516420364,
+      "rewards/margins": 0.3064225912094116,
+      "rewards/rejected": -0.38852202892303467,
+      "step": 1984
     }
   ],
   "logging_steps": 1,

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33cc5549722e683d6baf191851872bb5933c9f126da3923b5041d213900c4aed
+oid sha256:068f9f17f37a0b6c5693d6feb99bb88153b523b8cdd47d5dc7a2c2ceae6f4d45
 size 5496