Training in progress, step 4000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c3f58047ad6c40a78585613022f4ed80be220bf6ec21185d07d2c8a498c4d81
+oid sha256:c3eb29474e73931dfcd91c2e79d9d8a79eabb6944e3d84da49f7f90f18979f71
 size 6710771584

last-checkpoint/pytorch_model-00001-of-00003.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c82a2e7c8ec4e63cb1fe1265398cbd375ba2db2c8e3306d8c4f71b7aee6a9650
+oid sha256:3667ce09a5903ad6e193c00d0ddb89daaea8db351add98c814da23dbafea530d
 size 4986145435

last-checkpoint/pytorch_model-00002-of-00003.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ffc65c9f1d6f86ab21566e966a8ee5f99bb6514bc9a833ae54f735ad7480417
+oid sha256:8a11d04976761ef8277f44a0f7c0e228b3cac969b10ee5c6b4de7b171c800cc9
 size 4985796040

last-checkpoint/pytorch_model-00003-of-00003.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23ce7fd3a2949f3878635d26b1caf6c1559d0813d1ebd24edc72fb58c2187e58
+oid sha256:2a4a2096581fc9ebadb506824fa7258859cda3d3274ad32362d499c3cb4e45b7
 size 3407874846

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b5bfe987daf29ca780f6f0f48d581306ba9822c23e7ac806d164694cc658fe0
+oid sha256:0592b22bbf2c12a8a47cfcd0fc1dc3c2494837461af5f07d2b2ae6a9fe0dfde8
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8d3846c8c4611a93d268239a3a849d9aeb15e0e658fb1d1fc10fa99b8483e00
+oid sha256:78d61582d761df7115a63fb89688f5937274cf5adaaf644e15aaa05c94e6b14d
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 1.1881097555160522,
   "best_model_checkpoint": "../facebook/nllb-200-3.3B-finetuned/checkpoint-1000",
-  "epoch": 3500.0,
+  "epoch": 4000.0,
   "eval_steps": 500,
-  "global_step": 3500,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -126,6 +126,23 @@
       "eval_samples_per_second": 0.289,
       "eval_steps_per_second": 0.144,
       "step": 3500
+    },
+    {
+      "epoch": 4000.0,
+      "grad_norm": 0.031249279156327248,
+      "learning_rate": 9.60461956544021e-07,
+      "loss": 0.0017,
+      "step": 4000
+    },
+    {
+      "epoch": 4000.0,
+      "eval_gen_len": 24.5,
+      "eval_loss": 1.4340063333511353,
+      "eval_rouge": 0.0357,
+      "eval_runtime": 6.9464,
+      "eval_samples_per_second": 0.288,
+      "eval_steps_per_second": 0.144,
+      "step": 4000
     }
   ],
   "logging_steps": 500,
@@ -133,7 +150,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5000,
   "save_steps": 500,
-  "total_flos": 2.2267930804224e+16,
+  "total_flos": 2.5449063776256e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null