Phase3 ckpt step 50

checkpoints/checkpoint-50/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c05602edd46e4cf53d8b2e2c7d9185cb9bb2dc759775117a05cd5c1589b2b27
+oid sha256:5f91cce9515723b65c6fafaacbdf9b3de96a50d42bbdebe0c8c7f30d1690f608
 size 119801528

checkpoints/checkpoint-50/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33c489ad20e393d48e76a172c78aafc073b3482e6f5405fe9250551604367237
+oid sha256:14ea8b43b3ff7d5fd2f18f25cd83613ed615bee210ff94290823a8bd573ab146
 size 61397701

checkpoints/checkpoint-50/trainer_state.json

@@ -11,44 +11,44 @@
   "log_history": [
     {
       "epoch": 0.019417475728155338,
-      "grad_norm": 0.5592218637466431,
+      "grad_norm": 0.6309196949005127,
       "learning_rate": 0.0,
-      "loss": 1.0037728548049927,
+      "loss": 1.007138967514038,
       "step": 1
     },
     {
       "epoch": 0.1941747572815534,
-      "grad_norm": 0.6074398159980774,
+      "grad_norm": 0.6809913516044617,
       "learning_rate": 9.524135262330098e-05,
-      "loss": 0.9681400722927518,
+      "loss": 0.9641060299343533,
       "step": 10
     },
     {
       "epoch": 0.3883495145631068,
-      "grad_norm": 0.6838647723197937,
+      "grad_norm": 0.6508110761642456,
       "learning_rate": 7.408768370508576e-05,
-      "loss": 0.9677780151367188,
+      "loss": 0.9679458618164063,
       "step": 20
     },
     {
       "epoch": 0.5825242718446602,
-      "grad_norm": 0.6627008318901062,
+      "grad_norm": 0.6529077887535095,
       "learning_rate": 4.373333832178478e-05,
-      "loss": 0.9933381080627441,
+      "loss": 0.956454086303711,
       "step": 30
     },
     {
       "epoch": 0.7766990291262136,
-      "grad_norm": 0.6596543192863464,
+      "grad_norm": 0.6825394630432129,
       "learning_rate": 1.5772644703565565e-05,
-      "loss": 0.9614946365356445,
+      "loss": 0.9437665939331055,
       "step": 40
     },
     {
       "epoch": 0.970873786407767,
-      "grad_norm": 0.6468517780303955,
+      "grad_norm": 0.6889418959617615,
       "learning_rate": 8.856374635655695e-07,
-      "loss": 0.9476834297180176,
+      "loss": 0.9414969444274902,
       "step": 50
     }
   ],
@@ -69,7 +69,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6663849985376256.0,
+  "total_flos": 6649445008539648.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

checkpoints/checkpoint-50/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6ee30206dd9557cdb72e96f4e6d60f78dc1179da2c42c668f6f07ff70838e45
+oid sha256:d7eb50cc13c649388d04f9d19ca831e6db6e0f12509fe17a7c1bb09fd8072213
 size 5777