Training in progress, epoch 0

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_20/checkpoint-280",
+  "_name_or_path": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_sfd_bp_10/checkpoint-280",
   "architectures": [
     "LongT5ForConditionalGeneration"
   ],
@@ -29,7 +29,7 @@
   "relative_attention_num_buckets": 32,
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.34.1",
+  "transformers_version": "4.36.2",
   "use_cache": true,
   "vocab_size": 32128
 }

tokenizer.json

@@ -949,7 +949,8 @@
       {
         "type": "Metaspace",
         "replacement": "▁",
-        "add_prefix_space": true
+        "add_prefix_space": true,
+        "prepend_scheme": "always"
       }
     ]
   },
@@ -1010,7 +1011,8 @@
   "decoder": {
     "type": "Metaspace",
     "replacement": "▁",
-    "add_prefix_space": true
+    "add_prefix_space": true,
+    "prepend_scheme": "always"
   },
   "model": {
     "type": "Unigram",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7515f0d88ae11b45c396561915aafc121149dd3d4e0eb155075e83079c33ee6c
-size 4792
+oid sha256:2ebc46bf5b4a049340d2ad833de6d7a6690dc5ec53c0a9f478c89a402e7378fe
+size 4984