First model

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 4.71,
+    "eval_loss": 1.5495115518569946,
+    "eval_runtime": 200.1691,
+    "eval_samples": 181300,
+    "eval_samples_per_second": 905.734,
+    "eval_steps_per_second": 3.542,
+    "perplexity": 4.709169435672752,
+    "train_loss": 0.5639808364868164,
+    "train_runtime": 35161.1614,
+    "train_samples": 4349088,
+    "train_samples_per_second": 582.461,
+    "train_steps_per_second": 0.284
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "finiteautomata/robertuito-base-uncased",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 130,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.13.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 30000
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.71,
+    "eval_loss": 1.5495115518569946,
+    "eval_runtime": 200.1691,
+    "eval_samples": 181300,
+    "eval_samples_per_second": 905.734,
+    "eval_steps_per_second": 3.542,
+    "perplexity": 4.709169435672752
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e20de70f23ab1cb59069672cff7ffc92478f3246ec0faf58929c9149c284a6cb
+size 435341035

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": "models/twerto-base-uncased/special_tokens_map.json", "name_or_path": "finiteautomata/robertuito-base-uncased", "tokenizer_class": "PreTrainedTokenizerFast"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.71,
+    "train_loss": 0.5639808364868164,
+    "train_runtime": 35161.1614,
+    "train_samples": 4349088,
+    "train_samples_per_second": 582.461,
+    "train_steps_per_second": 0.284
+}