End of training

README.md

@@ -32,15 +32,13 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0005
+- learning_rate: 5e-05
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- num_epochs: 300
+- lr_scheduler_type: linear
+- num_epochs: 3.0
 
 ### Training results
 

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31aa75174f1171830bbbd66a21b8d5a782f80e1a78e5d0006d746d19af98d641
-size 1324917277
+oid sha256:7d80ed7d49f311a809b2250816134017b2f8fd2fb84edadede4687d22562508f
+size 1324911837

special_tokens_map.json

@@ -1,6 +1,6 @@
 {
   "bos_token": "</s>",
   "eos_token": "</s>",
-  "pad_token": "</s>",
+  "pad_token": "<pad>",
   "unk_token": "</s>"
 }

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

tokenizer_config.json

@@ -25,7 +25,7 @@
   "eos_token": "</s>",
   "errors": "replace",
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "</s>",
+  "pad_token": "<pad>",
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "</s>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a499fd307ebb77a86f6a84938c4bd727e699e35ea9c55d2a0139d41b74511ab
+oid sha256:a63dfdc47e8b387ef099ad7587a389f1b89c0336fb4de6bc94b0354187b16f0a
 size 4091