Training in progress, step 3000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6594507d2093f7d04ea48dba9b17f445e9541e34c9ab8598b25c91de8e5ab89
+oid sha256:4c73bd4e1481de3d4144f2e9366506214406383a3782a6f05e0935ebd0f8af9e
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe7a50c86b765550303567a7c71dc3287d2a1793b843374f3662606fefc4c7b7
+oid sha256:16f77093530f639a4af7dd28570340d9f1a541bf5654172b51c5775871c3341b
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e8aa2535c3a1ec264ce2c31a7ac416d4861404181dd39ae479182880849bcf2
+oid sha256:f234102d24bc6a02d177cba50bef1f62b316dd5f3618069f330bf9b4062dafbc
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8452e8180915c2a4933789804dc213986b2971ca2876fd0285cc8294b6a4c56
+oid sha256:a0ca419b875f31d2ca3779860c469be02e54b8cb9c32f39af28ad06b56000aaf
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3997121751308441,
-  "best_model_checkpoint": "mikhail-panzo/ceb_b64_le5_s8000/checkpoint-2500",
-  "epoch": 98.03921568627452,
+  "best_metric": 0.395207017660141,
+  "best_model_checkpoint": "mikhail-panzo/ceb_b64_le5_s8000/checkpoint-3000",
+  "epoch": 117.6470588235294,
   "eval_steps": 500,
-  "global_step": 2500,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -397,6 +397,84 @@
       "eval_samples_per_second": 28.109,
       "eval_steps_per_second": 3.592,
       "step": 2500
+    },
+    {
+      "epoch": 100.0,
+      "grad_norm": 2.315229892730713,
+      "learning_rate": 9.085e-06,
+      "loss": 0.4295,
+      "step": 2550
+    },
+    {
+      "epoch": 101.96078431372548,
+      "grad_norm": 1.0224589109420776,
+      "learning_rate": 9.001666666666667e-06,
+      "loss": 0.4212,
+      "step": 2600
+    },
+    {
+      "epoch": 103.92156862745098,
+      "grad_norm": 1.389236330986023,
+      "learning_rate": 8.918333333333334e-06,
+      "loss": 0.4215,
+      "step": 2650
+    },
+    {
+      "epoch": 105.88235294117646,
+      "grad_norm": 0.9601902961730957,
+      "learning_rate": 8.836666666666668e-06,
+      "loss": 0.4271,
+      "step": 2700
+    },
+    {
+      "epoch": 107.84313725490196,
+      "grad_norm": 0.8070810437202454,
+      "learning_rate": 8.753333333333333e-06,
+      "loss": 0.4269,
+      "step": 2750
+    },
+    {
+      "epoch": 109.80392156862744,
+      "grad_norm": 1.3389467000961304,
+      "learning_rate": 8.67e-06,
+      "loss": 0.4209,
+      "step": 2800
+    },
+    {
+      "epoch": 111.76470588235294,
+      "grad_norm": 1.0083783864974976,
+      "learning_rate": 8.586666666666667e-06,
+      "loss": 0.4143,
+      "step": 2850
+    },
+    {
+      "epoch": 113.72549019607843,
+      "grad_norm": 1.390769600868225,
+      "learning_rate": 8.503333333333334e-06,
+      "loss": 0.4208,
+      "step": 2900
+    },
+    {
+      "epoch": 115.68627450980392,
+      "grad_norm": 1.0967605113983154,
+      "learning_rate": 8.42e-06,
+      "loss": 0.4127,
+      "step": 2950
+    },
+    {
+      "epoch": 117.6470588235294,
+      "grad_norm": 0.9284217357635498,
+      "learning_rate": 8.336666666666668e-06,
+      "loss": 0.4166,
+      "step": 3000
+    },
+    {
+      "epoch": 117.6470588235294,
+      "eval_loss": 0.395207017660141,
+      "eval_runtime": 6.3246,
+      "eval_samples_per_second": 28.46,
+      "eval_steps_per_second": 3.637,
+      "step": 3000
     }
   ],
   "logging_steps": 50,
@@ -416,7 +494,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.7204142566350376e+16,
+  "total_flos": 3.26453987568258e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null