Training in progress, step 10000, checkpoint

last-checkpoint/generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "bos_token_id": 0,
   "decoder_start_token_id": 0,
   "eos_token_id": 1,

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d89db5d06633ee35b29ce1cdcb2aa554e530e71c804f3829c99278901f56873
+oid sha256:f1144d55646fc5998952b92fbb22ebfdad6d9bfe6cbf4765e664b4587dd96128
 size 959732416

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:856fe61bcf3c3ab44324a4c7452f1e2417f81476343add229b792d2be1f5c90c
+oid sha256:5dc047b3cd666b27e148127d9b13556cf3973cacb8db040dd3559e43fa5214ff
 size 1915006400

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c110c6d831b996ab5e1cd8f94d51b06d224e5149023c78d224a7ceba5a4457c
+oid sha256:5ab51f4ca4a16a4bfb91c3d1be6507a2eaed5f11fdaba6a41f061d41eded8588
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a829688278493105074ae1f56d19475c0dd11acdbb38079764885645edeb6f52
+oid sha256:f07d7ed641fa38345ee876b00d140a22954bc1b2580281f386922507f95c1018
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.8385041086701325,
+  "epoch": 1.677008217340265,
   "eval_steps": 500,
-  "global_step": 5000,
+  "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -77,6 +77,89 @@
       "learning_rate": 4.161495891329868e-05,
       "loss": 0.1876,
       "step": 5000
+    },
+    {
+      "epoch": 0.9223545195371458,
+      "grad_norm": 0.5806877613067627,
+      "learning_rate": 4.0776454804628546e-05,
+      "loss": 0.1753,
+      "step": 5500
+    },
+    {
+      "epoch": 1.0,
+      "eval_gen_len": 100.0,
+      "eval_loss": 0.15864285826683044,
+      "eval_rouge1": 0.0,
+      "eval_rouge2": 0.0,
+      "eval_rougeL": 0.0,
+      "eval_rougeLsum": 0.0,
+      "eval_runtime": 1668.5377,
+      "eval_samples_per_second": 0.795,
+      "eval_steps_per_second": 0.397,
+      "step": 5963
+    },
+    {
+      "epoch": 1.006204930404159,
+      "grad_norm": 0.4943917691707611,
+      "learning_rate": 3.9937950695958413e-05,
+      "loss": 0.1552,
+      "step": 6000
+    },
+    {
+      "epoch": 1.0900553412711722,
+      "grad_norm": 1.4509881734848022,
+      "learning_rate": 3.909944658728828e-05,
+      "loss": 0.1244,
+      "step": 6500
+    },
+    {
+      "epoch": 1.1739057521381855,
+      "grad_norm": 0.4178345203399658,
+      "learning_rate": 3.826094247861814e-05,
+      "loss": 0.1326,
+      "step": 7000
+    },
+    {
+      "epoch": 1.2577561630051988,
+      "grad_norm": 0.7037866711616516,
+      "learning_rate": 3.7422438369948016e-05,
+      "loss": 0.1277,
+      "step": 7500
+    },
+    {
+      "epoch": 1.341606573872212,
+      "grad_norm": 0.868385910987854,
+      "learning_rate": 3.658393426127788e-05,
+      "loss": 0.1279,
+      "step": 8000
+    },
+    {
+      "epoch": 1.4254569847392253,
+      "grad_norm": 0.4815532863140106,
+      "learning_rate": 3.574543015260775e-05,
+      "loss": 0.1269,
+      "step": 8500
+    },
+    {
+      "epoch": 1.5093073956062386,
+      "grad_norm": 0.5484045147895813,
+      "learning_rate": 3.490692604393762e-05,
+      "loss": 0.1308,
+      "step": 9000
+    },
+    {
+      "epoch": 1.5931578064732517,
+      "grad_norm": 0.4542590081691742,
+      "learning_rate": 3.4068421935267486e-05,
+      "loss": 0.1259,
+      "step": 9500
+    },
+    {
+      "epoch": 1.677008217340265,
+      "grad_norm": 0.26205527782440186,
+      "learning_rate": 3.3229917826597354e-05,
+      "loss": 0.1117,
+      "step": 10000
     }
   ],
   "logging_steps": 500,
@@ -96,7 +179,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.80463891120128e+18,
+  "total_flos": 3.60909735851144e+18,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null