Training in progress, step 3000, checkpoint

last-checkpoint/config.json

@@ -19,7 +19,7 @@
   "num_hidden_layers": 12,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
-  "transformers_version": "4.57.1",
+  "transformers_version": "4.57.3",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 50265

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79cc233a6be9a69c6d79a84da18b8aa1c085a8ba9ed3fc676b30dd261b8b10c2
+oid sha256:0cc173eb2df771065213d42995586028dd871d15a4d0f971a3e924c00d05221b
 size 498612824

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c3f1315d4876eb442bc815dcd6cfd83cf26ce5d2ba594754037d92371255cf1
+oid sha256:14c16fca80bbfab50a9063aa5d2d7ac7fec1181f50a559b2ba121ecfee922439
 size 997345931

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8cbb9d20ae0e060525bcac1b70551f336242a3392c3ba0392910b81075e6dc33
+oid sha256:abf2b0fc64c4be716a75868bdb69c6ff3620ba2a21085b2c861288da2529d5e7
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
+oid sha256:5ac1c46a2776d12775d23d0f587efc112188137ce2140da35bc15d301c9f620e
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38a85adeb1e28a9f8d2b9e5eda88bb40a25fff60709bb465ee86ac35b1f82adc
+oid sha256:7fe4d93dfc6f647f747e92a74d9d5986e59893a476d774e2205030dc67d6ff6d
 size 1465

last-checkpoint/trainer_state.json

@@ -1,63 +1,86 @@
 {
   "best_global_step": 1000,
-  "best_metric": 0.08424179255962372,
+  "best_metric": 0.10802757740020752,
   "best_model_checkpoint": "./training_output/checkpoint-1000",
-  "epoch": 0.25,
+  "epoch": 0.15,
   "eval_steps": 1000,
-  "global_step": 2000,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0625,
-      "grad_norm": 33.47822570800781,
-      "learning_rate": 1.8752500000000004e-05,
-      "loss": 0.1408,
+      "epoch": 0.025,
+      "grad_norm": 0.8834348917007446,
+      "learning_rate": 1.9501000000000002e-05,
+      "loss": 0.1437,
       "step": 500
     },
     {
-      "epoch": 0.125,
-      "grad_norm": 66.1669692993164,
-      "learning_rate": 1.7502500000000004e-05,
-      "loss": 0.1244,
+      "epoch": 0.05,
+      "grad_norm": 9.513919830322266,
+      "learning_rate": 1.9001e-05,
+      "loss": 0.1085,
       "step": 1000
     },
     {
-      "epoch": 0.125,
-      "eval_accuracy": 0.931,
-      "eval_loss": 0.08424179255962372,
-      "eval_runtime": 148.5886,
-      "eval_samples_per_second": 53.84,
-      "eval_steps_per_second": 3.365,
+      "epoch": 0.05,
+      "eval_accuracy": 0.91835,
+      "eval_loss": 0.10802757740020752,
+      "eval_runtime": 381.1922,
+      "eval_samples_per_second": 52.467,
+      "eval_steps_per_second": 3.279,
       "step": 1000
     },
     {
-      "epoch": 0.1875,
-      "grad_norm": 40.86865234375,
-      "learning_rate": 1.62525e-05,
-      "loss": 0.083,
+      "epoch": 0.075,
+      "grad_norm": 0.013394818641245365,
+      "learning_rate": 1.8501e-05,
+      "loss": 0.0965,
       "step": 1500
     },
     {
-      "epoch": 0.25,
-      "grad_norm": 0.046407029032707214,
-      "learning_rate": 1.5002500000000002e-05,
-      "loss": 0.0851,
+      "epoch": 0.1,
+      "grad_norm": 0.017180055379867554,
+      "learning_rate": 1.8001000000000003e-05,
+      "loss": 0.0716,
       "step": 2000
     },
     {
-      "epoch": 0.25,
-      "eval_accuracy": 0.7715,
-      "eval_loss": 0.20412930846214294,
-      "eval_runtime": 149.0575,
-      "eval_samples_per_second": 53.671,
-      "eval_steps_per_second": 3.354,
+      "epoch": 0.1,
+      "eval_accuracy": 0.8473,
+      "eval_loss": 0.25237375497817993,
+      "eval_runtime": 381.1616,
+      "eval_samples_per_second": 52.471,
+      "eval_steps_per_second": 3.279,
       "step": 2000
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 0.017619747668504715,
+      "learning_rate": 1.7501e-05,
+      "loss": 0.0658,
+      "step": 2500
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.527113676071167,
+      "learning_rate": 1.7001000000000002e-05,
+      "loss": 0.0615,
+      "step": 3000
+    },
+    {
+      "epoch": 0.15,
+      "eval_accuracy": 0.9299,
+      "eval_loss": 0.11818733811378479,
+      "eval_runtime": 381.4889,
+      "eval_samples_per_second": 52.426,
+      "eval_steps_per_second": 3.277,
+      "step": 3000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 8000,
+  "max_steps": 20000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 1000,
@@ -73,7 +96,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2104888442880000.0,
+  "total_flos": 3157332664320000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d1dcf0aa731896a45dc92ce645a0147a079fc2db258f2967bafe27b497eba6b
+oid sha256:4672bef08597eaac4166af409003080f023f24990f8f58c185b2b990119dc93e
 size 5841