Training in progress, step 7000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44e5337b3e0214b4729f349032988a9b03806846f7f4bd6d268aef041b459c52
+oid sha256:a082886572e13067532489a5fc1dcb98c513570df4676ae62b39ded1c25bb7e4
 size 441688704

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59fe147cb04454b19d7862577b716d215f62ea0d638e3a53e231ac5e1783553f
+oid sha256:72a510a1be1e74db7c7d7a99d3f3d239298e6879c992b42b0e0f01ea7eacdb9b
 size 883473803

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4eb312d65febc64758e9d1fe191e5c3c93cd8ec224be66347404eb3503c7f723
+oid sha256:5a3c0e5c96be49e3d0f56967ef3a66294b8033a429d896ba1c57076548eb65a4
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bd42e357d6b98cb9bff543c6a4f4b9f46ab579a827a6b5d2df5dd5d499424c0
+oid sha256:6a68ceae7165f4c88c8c4809f6ef958049408efc4229565cd13a1f7800728538
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bfe8ee0b406625e2e6930efb21c3d4599f197be5d04a29012b92920efae4631
+oid sha256:d119247b2cee9414eefc7d0552fb04c1f86b7b19685ed262e23f9b9c7cf23594
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.7647058823529411,
+  "epoch": 0.8235294117647058,
   "eval_steps": 500,
-  "global_step": 6500,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1022,6 +1022,84 @@
       "eval_samples_per_second": 49.549,
       "eval_steps_per_second": 1.55,
       "step": 6500
+    },
+    {
+      "epoch": 0.7705882352941177,
+      "grad_norm": 0.3573964536190033,
+      "learning_rate": 0.00011476470588235294,
+      "loss": 3.5397024536132813,
+      "step": 6550
+    },
+    {
+      "epoch": 0.7764705882352941,
+      "grad_norm": 0.3719513416290283,
+      "learning_rate": 0.00011182352941176472,
+      "loss": 3.5319180297851562,
+      "step": 6600
+    },
+    {
+      "epoch": 0.7823529411764706,
+      "grad_norm": 0.3488079905509949,
+      "learning_rate": 0.00010888235294117648,
+      "loss": 3.525909118652344,
+      "step": 6650
+    },
+    {
+      "epoch": 0.788235294117647,
+      "grad_norm": 0.378339022397995,
+      "learning_rate": 0.00010594117647058823,
+      "loss": 3.5297225952148437,
+      "step": 6700
+    },
+    {
+      "epoch": 0.7941176470588235,
+      "grad_norm": 0.3940635919570923,
+      "learning_rate": 0.000103,
+      "loss": 3.544861145019531,
+      "step": 6750
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.36259767413139343,
+      "learning_rate": 0.00010005882352941177,
+      "loss": 3.5146881103515626,
+      "step": 6800
+    },
+    {
+      "epoch": 0.8058823529411765,
+      "grad_norm": 0.35855287313461304,
+      "learning_rate": 9.711764705882353e-05,
+      "loss": 3.4209246826171875,
+      "step": 6850
+    },
+    {
+      "epoch": 0.8117647058823529,
+      "grad_norm": 0.3636466860771179,
+      "learning_rate": 9.417647058823529e-05,
+      "loss": 3.268857421875,
+      "step": 6900
+    },
+    {
+      "epoch": 0.8176470588235294,
+      "grad_norm": 0.36832499504089355,
+      "learning_rate": 9.123529411764706e-05,
+      "loss": 3.299452209472656,
+      "step": 6950
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.3607626259326935,
+      "learning_rate": 8.829411764705883e-05,
+      "loss": 3.5039678955078126,
+      "step": 7000
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "eval_loss": 3.3713526725769043,
+      "eval_runtime": 176.7042,
+      "eval_samples_per_second": 49.195,
+      "eval_steps_per_second": 1.539,
+      "step": 7000
     }
   ],
   "logging_steps": 50,
@@ -1041,7 +1119,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.08697485312e+17,
+  "total_flos": 1.17058830336e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null