Training in progress epoch 16

README.md

@@ -13,9 +13,9 @@ 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:
-- Train Loss: 1.2264
-- Validation Loss: 1.0297
-- Epoch: 15
+- Train Loss: 1.2292
+- Validation Loss: 1.0824
+- Epoch: 16
 
 ## Model description
 
@@ -41,17 +41,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Epoch |
 |:----------:|:---------------:|:-----:|
-| 1.2445     | 1.0249          | 5     |
-| 1.2386     | 1.0253          | 6     |
-| 1.2438     | 1.0271          | 7     |
-| 1.2405     | 1.0280          | 8     |
-| 1.2384     | 1.0276          | 9     |
-| 1.2351     | 1.0297          | 10    |
-| 1.2332     | 1.0295          | 11    |
-| 1.2361     | 1.0295          | 12    |
-| 1.2303     | 1.0307          | 13    |
-| 1.2253     | 1.0312          | 14    |
-| 1.2264     | 1.0297          | 15    |
+| 1.2292     | 1.0824          | 16    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1759a19cada8cb3acd15f988fba3dc254322a9bad1c1051587f14712dc3d1caf
+oid sha256:783af7d97a235cb94cb3870acbc91992f3222e81e4568b9fc8a286c5fcc12e4b
 size 373902664

tokenizer.json

@@ -2,12 +2,14 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 512,
+    "max_length": 128,
     "strategy": "LongestFirst",
     "stride": 0
   },
   "padding": {
-    "strategy": "BatchLongest",
+    "strategy": {
+      "Fixed": 128
+    },
     "direction": "Right",
     "pad_to_multiple_of": null,
     "pad_id": 0,