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: 0.8802
-- eval_runtime: 1993.2844
-- eval_samples_per_second: 111.933
-- eval_steps_per_second: 1.749
-- epoch: 18.33
-- step: 1150000
+- eval_loss: 0.8813
+- eval_runtime: 2003.661
+- eval_samples_per_second: 111.354
+- eval_steps_per_second: 1.74
+- epoch: 19.12
+- step: 1200000
 
 ## Model description
 

logs/events.out.tfevents.1705461624.398ab36d7f47.26.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c50dde48c1d0bbda18b53406ce302c9be5e445331386f9698f7961b5f80b21b
+size 5099

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eeff65b8f9080bd84229af9009c4f3ca96ab5b56561ffa57e8bbe58a8cc399fc
+oid sha256:c66f8cbaa7b4022b34fab9545aac3eee47ad1e7bfad0a0c3f863b03566a5e24d
 size 497918592

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 18.326401172889675,
+  "epoch": 19.123201223884877,
   "eval_steps": 50000,
-  "global_step": 1150000,
+  "global_step": 1200000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -329,6 +329,20 @@
       "eval_samples_per_second": 111.933,
       "eval_steps_per_second": 1.749,
       "step": 1150000
+    },
+    {
+      "epoch": 19.12,
+      "learning_rate": 4.992034549250211e-05,
+      "loss": 0.8399,
+      "step": 1200000
+    },
+    {
+      "epoch": 19.12,
+      "eval_loss": 0.8813353180885315,
+      "eval_runtime": 2003.661,
+      "eval_samples_per_second": 111.354,
+      "eval_steps_per_second": 1.74,
+      "step": 1200000
     }
   ],
   "logging_steps": 50000,
@@ -336,7 +350,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 50000,
-  "total_flos": 2.40388081532928e+18,
+  "total_flos": 2.50839730151424e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null