Training in progress, step 500

config.json

@@ -1,27 +1,24 @@
 {
-  "_name_or_path": "roberta-base",
+  "_name_or_path": "distilbert-base-uncased",
+  "activation": "gelu",
   "architectures": [
-    "RobertaForMaskedLM"
+    "DistilBertForMaskedLM"
   ],
-  "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": 0,
-  "classifier_dropout": null,
-  "eos_token_id": 2,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
-  "layer_norm_eps": 1e-05,
-  "max_position_embeddings": 514,
-  "model_type": "roberta",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
-  "pad_token_id": 1,
-  "position_embedding_type": "absolute",
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.19.4",
-  "type_vocab_size": 1,
-  "use_cache": true,
-  "vocab_size": 2500
+  "transformers_version": "4.16.2",
+  "vocab_size": 20000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:913f94518d1932cc1383cb260e5575bb3395b1f43198f80c3b7980256145e1bc
-size 351936555
+oid sha256:fe6414f4a2f3f556ddf40685991363e3db8108fa538b26b98638ee46442ce626
+size 235600854

runs/Jul12_10-48-03_lu-Aspire-T3-715A/1657637452.751271/events.out.tfevents.1657637452.lu-Aspire-T3-715A.22494.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e41e423bf3d36b50c4d3f0bb0ff1144629ca7a91bc12df60d97848014b4d3746
+size 4556

runs/Jul12_10-48-03_lu-Aspire-T3-715A/events.out.tfevents.1657637452.lu-Aspire-T3-715A.22494.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de917ebfb950b743fc5950d1c843b726a3d0709dfcb67bb45ccc17dade6228c0
+size 40

runs/Jul12_10-56-34_lu-Aspire-T3-715A/1657637800.2452366/events.out.tfevents.1657637800.lu-Aspire-T3-715A.24933.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726a6526b63f533a6529a6e92cadf6eead733213169d655eb3ebecf1ccd63dd4
+size 4556

runs/Jul12_10-56-34_lu-Aspire-T3-715A/events.out.tfevents.1657637800.lu-Aspire-T3-715A.24933.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14a6a32091a16dd37e8111554e9eb22694eee25062178ca5a9f9aef1d9f1dc8c
+size 3308

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b584a455f37b82deb1c1174b3f763d37511160c3ca8f9be5d84d575bf4afd7d
-size 3183
+oid sha256:5b7573152aabe4be36a7972233721c7a7e023143853fae96afe470208b36862c
+size 2741