Training in progress, epoch 2

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77ee03a1d0311a29793f120677b5bfc69e1e8781bc96853c8e9c88c8166c16e4
+oid sha256:1351bd3bf566cfe6d40151bfcbbf3bccee762d4f23ee53f6d2d73d2a5f861a64
 size 1115579397

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64abc4de3991cbdfd5ff0c316e2fac01273b5b67804a7c101f49a609279180f1
+oid sha256:1e95e8c522521c337928967e212665cfb3206fe4eb5ec06a2bb1372b913d6fc0
 size 557971229

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a45b7b77b12af6e89720c582cf4778d8f486d6946f3e82947292c4d5038406f7
+oid sha256:154076bccd841f8ac822307d4f056a539bae87f1cf1cd71150f9a588ea1896b8
 size 15597

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03c25bfe5cf11a212342871a1fefb1e71d639bd1297dd40b8d896a9d5808de6b
+oid sha256:6fe54476cab8566f02dbdb424711234a968a086cef865765d8cdb7f900b7dbb6
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0801195a6d147316da17a810626a833a27ae6da4b2ee1058dbb32488537477c
+oid sha256:81fe517ff574cd498c420225231c23c454eca4639d3d380b67b10c0d2f673ad4
 size 627

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
-  "global_step": 5534,
+  "epoch": 2.0,
+  "global_step": 11068,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -98,11 +98,116 @@
       "learning_rate": 0.00015035236718467654,
       "loss": 2.0776,
       "step": 5500
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.0001458348391760029,
+      "loss": 1.7518,
+      "step": 6000
+    },
+    {
+      "epoch": 1.08,
+      "eval_gen_len": 19.6188,
+      "eval_loss": 1.6170521974563599,
+      "eval_rouge1": 1.4387,
+      "eval_rouge2": 1.2076,
+      "eval_rougeL": 1.4167,
+      "eval_rougeLsum": 1.4222,
+      "eval_runtime": 3886.2514,
+      "eval_samples_per_second": 7.119,
+      "eval_steps_per_second": 0.356,
+      "step": 6000
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00014131731116732925,
+      "loss": 1.7539,
+      "step": 6500
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00013680881821467292,
+      "loss": 1.7379,
+      "step": 7000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0001322912902059993,
+      "loss": 1.7458,
+      "step": 7500
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00012777376219732562,
+      "loss": 1.7498,
+      "step": 8000
+    },
+    {
+      "epoch": 1.45,
+      "eval_gen_len": 19.549,
+      "eval_loss": 1.5690149068832397,
+      "eval_rouge1": 1.6999,
+      "eval_rouge2": 1.4397,
+      "eval_rougeL": 1.6688,
+      "eval_rougeLsum": 1.6791,
+      "eval_runtime": 3827.5328,
+      "eval_samples_per_second": 7.228,
+      "eval_steps_per_second": 0.362,
+      "step": 8000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00012325623418865198,
+      "loss": 1.7506,
+      "step": 8500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00011873870617997833,
+      "loss": 1.7461,
+      "step": 9000
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00011422117817130467,
+      "loss": 1.7558,
+      "step": 9500
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00010970365016263103,
+      "loss": 1.7326,
+      "step": 10000
+    },
+    {
+      "epoch": 1.81,
+      "eval_gen_len": 19.4797,
+      "eval_loss": 1.5457514524459839,
+      "eval_rouge1": 1.6783,
+      "eval_rouge2": 1.3966,
+      "eval_rougeL": 1.6484,
+      "eval_rougeLsum": 1.658,
+      "eval_runtime": 3817.3069,
+      "eval_samples_per_second": 7.247,
+      "eval_steps_per_second": 0.363,
+      "step": 10000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00010518612215395735,
+      "loss": 1.7658,
+      "step": 10500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00010067762920130106,
+      "loss": 1.7348,
+      "step": 11000
     }
   ],
   "max_steps": 22136,
   "num_train_epochs": 4,
-  "total_flos": 6.744738452594688e+16,
+  "total_flos": 1.348859886087168e+17,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64abc4de3991cbdfd5ff0c316e2fac01273b5b67804a7c101f49a609279180f1
+oid sha256:1e95e8c522521c337928967e212665cfb3206fe4eb5ec06a2bb1372b913d6fc0
 size 557971229