Training in progress, step 8000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95b216ca2f0f331cd0fb1b5060aae7303e018fd811a626e11c3f50b85f075124
+oid sha256:d70198e690533c8705cb606a2f8554e9c88db3919bc157b0b85cfd40b9163cdf
 size 441688704

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:730fa392e5f88931f6f53528627829ab93c230c150c53ca715559afeb980893d
+oid sha256:f09de07399f3f34149aed6397af707e189d23d41a5d5df6968ab6e71d7d2629a
 size 883473803

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9ccaae7b7094e3f6ae09519a1d8593eea1044e5a091df08ba92284c2a368db5a
+oid sha256:ff6f17e3eb221fe462edccd50497bde0f3a71571ed479a52a41576181171c931
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:67e9e63342c9fc618c8284b5217646e8560302dc68c6c23a6c0fc311b7d330bf
+oid sha256:a95554cb3d06aec9d05e31c8207a60744e94359015aa846162464c9cdf88bcec
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6847feb481bac2207bbdccaefe424c80456cc08ed8d62eb651573dfcd4ddaea
+oid sha256:fef1bbe15ebf93cd14ffe46ac40b68f1ba1e52a58545c7076dbc53cd30fcde1a
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.8823529411764706,
+  "epoch": 0.9411764705882353,
   "eval_steps": 500,
-  "global_step": 7500,
+  "global_step": 8000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1178,6 +1178,84 @@
       "eval_samples_per_second": 49.467,
       "eval_steps_per_second": 1.548,
       "step": 7500
+    },
+    {
+      "epoch": 0.888235294117647,
+      "grad_norm": 0.3486090898513794,
+      "learning_rate": 5.5941176470588236e-05,
+      "loss": 3.2706234741210936,
+      "step": 7550
+    },
+    {
+      "epoch": 0.8941176470588236,
+      "grad_norm": 0.3703632950782776,
+      "learning_rate": 5.3e-05,
+      "loss": 3.465188293457031,
+      "step": 7600
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.3537668287754059,
+      "learning_rate": 5.0058823529411766e-05,
+      "loss": 3.487032470703125,
+      "step": 7650
+    },
+    {
+      "epoch": 0.9058823529411765,
+      "grad_norm": 0.3689066171646118,
+      "learning_rate": 4.711764705882353e-05,
+      "loss": 3.5042953491210938,
+      "step": 7700
+    },
+    {
+      "epoch": 0.9117647058823529,
+      "grad_norm": 0.3667464256286621,
+      "learning_rate": 4.4176470588235296e-05,
+      "loss": 3.527998962402344,
+      "step": 7750
+    },
+    {
+      "epoch": 0.9176470588235294,
+      "grad_norm": 0.36605432629585266,
+      "learning_rate": 4.123529411764706e-05,
+      "loss": 3.484926452636719,
+      "step": 7800
+    },
+    {
+      "epoch": 0.9235294117647059,
+      "grad_norm": 0.3601936101913452,
+      "learning_rate": 3.8294117647058826e-05,
+      "loss": 3.4993707275390626,
+      "step": 7850
+    },
+    {
+      "epoch": 0.9294117647058824,
+      "grad_norm": 0.35972827672958374,
+      "learning_rate": 3.535294117647059e-05,
+      "loss": 3.5094595336914063,
+      "step": 7900
+    },
+    {
+      "epoch": 0.9352941176470588,
+      "grad_norm": 0.4135211706161499,
+      "learning_rate": 3.2411764705882356e-05,
+      "loss": 3.512149658203125,
+      "step": 7950
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.33782321214675903,
+      "learning_rate": 2.9470588235294117e-05,
+      "loss": 3.491231689453125,
+      "step": 8000
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "eval_loss": 3.343282461166382,
+      "eval_runtime": 176.7083,
+      "eval_samples_per_second": 49.194,
+      "eval_steps_per_second": 1.539,
+      "step": 8000
     }
   ],
   "logging_steps": 50,
@@ -1197,7 +1275,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.2542017536e+17,
+  "total_flos": 1.33781520384e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null