Training in progress, step 500

config.json

@@ -1,7 +1,7 @@
 {
-  "_name_or_path": "KETI-AIR/ke-t5-base",
+  "_name_or_path": "KETI-AIR-Downstream/long-ke-t5-base-translation-aihub-ko2en",
   "architectures": [
-    "T5ForConditionalGeneration"
+    "LongT5ForConditionalGeneration"
   ],
   "d_ff": 2048,
   "d_kv": 64,
@@ -9,22 +9,27 @@
   "decoder_start_token_id": 0,
   "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
+  "encoder_attention_type": "transient-global",
   "eos_token_id": 1,
   "feed_forward_proj": "gated-gelu",
+  "global_block_size": 16,
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
   "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
-  "model_type": "t5",
-  "n_positions": 512,
+  "local_radius": 127,
+  "model_type": "longt5",
+  "n_positions": 4096,
   "num_decoder_layers": 12,
   "num_heads": 12,
   "num_layers": 12,
+  "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
   "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.29.2",
   "use_cache": true,
-  "vocab_size": 64128
+  "vocab_size": 64100
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d8128c41299218f49a78133b2bd80b2d86b3981c9684aba021768efbc142cb4
-size 990015473
+oid sha256:45bf7850572db8b6d30f67ef166ae4a161ef7e68957644f857ee14efcbca564e
+size 1186888873

runs/Jun03_12-29-09_bd31fece64bd/1685795352.0320435/events.out.tfevents.1685795352.bd31fece64bd.168.18

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29a1be0d39441bb4c7b8165db0c1f75bd34fc103595a2259aca4c3b73b4e9bd4
+size 6237

runs/Jun03_12-29-09_bd31fece64bd/events.out.tfevents.1685795352.bd31fece64bd.168.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1929c10eec88817c94a738bae7f2dad3e1d93c7068b3a4f9f2cb3397e0c0aee2
+size 4496

special_tokens_map.json

@@ -101,6 +101,7 @@
     "<extra_id_98>",
     "<extra_id_99>"
   ],
+  "bos_token": "<s>",
   "eos_token": "</s>",
   "pad_token": "<pad>",
   "unk_token": "<unk>"

spiece.model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc595fcd800672d6ef0e2aea5ba0f1dc8826471f6f40a87b70619dcc60ddd4a
-size 1466734
+oid sha256:8084f94443e97fa1305bf48cf532a7ced0631070f2f570c156bfa1b3c72913dd
+size 1592561

tokenizer_config.json

@@ -101,10 +101,11 @@
     "<extra_id_98>",
     "<extra_id_99>"
   ],
+  "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "extra_ids": 100,
-  "model_max_length": 512,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "sp_model_kwargs": {},
   "tokenizer_class": "T5Tokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08918668d70979634de202aba82815bc709fcd24e530bb2b44f5061b021a3097
+oid sha256:dd076251275dc22ce8d9869bc80b44d71fbbaad521a7554719724556ef911cc3
 size 4091