Training in progress, step 500

config.json

@@ -7,11 +7,13 @@
   "d_kv": 64,
   "d_model": 768,
   "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
   "eos_token_id": 1,
   "feed_forward_proj": "gated-gelu",
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
+  "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
   "model_type": "mt5",
   "num_decoder_layers": 12,
@@ -24,7 +26,7 @@
   "tie_word_embeddings": false,
   "tokenizer_class": "T5Tokenizer",
   "torch_dtype": "float32",
-  "transformers_version": "4.19.2",
+  "transformers_version": "4.20.1",
   "use_cache": true,
   "vocab_size": 250112
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ec47c369e504e6da6f2fc57a4b58b2a66106c8195371541035a4320b56a6c78
-size 2329700301
+oid sha256:6c5b3daae9635191eca600565f9c6a178d93042b56a24c065287a868c27c3122
+size 2329700173

special_tokens_map.json

@@ -1 +1,5 @@
-{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
+{
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -1 +1,12 @@
-{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "sp_model_kwargs": {}, "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276", "tokenizer_file": null, "name_or_path": "google/mt5-base", "tokenizer_class": "T5Tokenizer"}
+{
+  "additional_special_tokens": null,
+  "eos_token": "</s>",
+  "extra_ids": 0,
+  "name_or_path": "google/mt5-base",
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276",
+  "tokenizer_class": "T5Tokenizer",
+  "tokenizer_file": null,
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ab1298028ffa290c3e98ffc7993f0ffdac78a623375a8ebb51c00baea565338
-size 3247
+oid sha256:5269fc4293d10045b283c1ed6ef0f4a2dcf8e8ea78882c76424f0c55ef7f560e
+size 3375