End of training

README.md

@@ -13,12 +13,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.8800
-- eval_runtime: 2021.1167
-- eval_samples_per_second: 110.392
-- eval_steps_per_second: 1.725
-- epoch: 23.11
-- step: 1450000
+- eval_loss: 0.8771
+- eval_runtime: 1995.2079
+- eval_samples_per_second: 111.825
+- eval_steps_per_second: 1.748
+- epoch: 23.9
+- step: 1500000
 
 ## Model description
 

logs/events.out.tfevents.1706094708.9715be663057.26.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cde1941b10e53f17b7eb8c8bf5fba2d595a6f7a7d12df24f282efe7dacfb185b
+size 5099

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1dbddb86f969282280a65ed4dd2012219eb863ac99138ee0331a62ace913f664
+oid sha256:6f0541783e22d8ff3ac5569ac1df2fbf083c4b29926687a32b871ac59ce20dc7
 size 497918592

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 23.107201478860894,
+  "epoch": 23.9040015298561,
   "eval_steps": 50000,
-  "global_step": 1450000,
+  "global_step": 1500000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -413,6 +413,20 @@
       "eval_samples_per_second": 110.392,
       "eval_steps_per_second": 1.725,
       "step": 1450000
+    },
+    {
+      "epoch": 23.9,
+      "learning_rate": 4.9920348679702315e-05,
+      "loss": 0.8277,
+      "step": 1500000
+    },
+    {
+      "epoch": 23.9,
+      "eval_loss": 0.8771235346794128,
+      "eval_runtime": 1995.2079,
+      "eval_samples_per_second": 111.825,
+      "eval_steps_per_second": 1.748,
+      "step": 1500000
     }
   ],
   "logging_steps": 50000,
@@ -420,7 +434,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 50000,
-  "total_flos": 3.03098005905408e+18,
+  "total_flos": 3.13549687185408e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null