Training in progress, step 500

config.json

@@ -1,60 +1,35 @@
 {
-  "_name_or_path": "t5-small",
+  "_name_or_path": "Stancld/longt5-tglobal-large-16384-pubmed-3k_steps",
   "architectures": [
-    "T5ForConditionalGeneration"
+    "LongT5ForConditionalGeneration"
   ],
-  "d_ff": 2048,
+  "d_ff": 2816,
   "d_kv": 64,
-  "d_model": 512,
+  "d_model": 1024,
   "decoder_start_token_id": 0,
-  "dense_act_fn": "relu",
+  "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
+  "encoder_attention_type": "transient-global",
   "eos_token_id": 1,
-  "feed_forward_proj": "relu",
+  "feed_forward_proj": "gated-gelu",
+  "global_block_size": 16,
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
-  "is_gated_act": false,
+  "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
-  "model_type": "t5",
-  "n_positions": 512,
-  "num_decoder_layers": 6,
-  "num_heads": 8,
-  "num_layers": 6,
+  "local_radius": 127,
+  "model_type": "longt5",
+  "n_positions": 4096,
+  "num_decoder_layers": 24,
+  "num_heads": 16,
+  "num_layers": 24,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
   "relative_attention_num_buckets": 32,
-  "task_specific_params": {
-    "summarization": {
-      "early_stopping": true,
-      "length_penalty": 2.0,
-      "max_length": 200,
-      "min_length": 30,
-      "no_repeat_ngram_size": 3,
-      "num_beams": 4,
-      "prefix": "summarize: "
-    },
-    "translation_en_to_de": {
-      "early_stopping": true,
-      "max_length": 300,
-      "num_beams": 4,
-      "prefix": "translate English to German: "
-    },
-    "translation_en_to_fr": {
-      "early_stopping": true,
-      "max_length": 300,
-      "num_beams": 4,
-      "prefix": "translate English to French: "
-    },
-    "translation_en_to_ro": {
-      "early_stopping": true,
-      "max_length": 300,
-      "num_beams": 4,
-      "prefix": "translate English to Romanian: "
-    }
-  },
-  "torch_dtype": "float32",
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
   "transformers_version": "4.28.0",
   "use_cache": true,
-  "vocab_size": 32128
+  "vocab_size": 32100
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d14ec456ceedd2d11fadaceae19a8ec72672405b6d0bfb11be5dc6763fadc60
-size 242070267
+oid sha256:cea6db3e053edc3ae3a259f1976a90e9c54f518978f8e33c2f2a25c2529bf0d7
+size 1566435763

tokenizer_config.json

@@ -104,7 +104,7 @@
   "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "extra_ids": 100,
-  "model_max_length": 512,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "tokenizer_class": "T5Tokenizer",
   "unk_token": "<unk>"

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b97471daf2b9cb5813193686432ef3e797cf95d32db5f16b6c83ad208b1b80c
+oid sha256:46379c3c32e5b2a101c8f931e5f840e82671c525c66201a05460731b4dc05623
 size 3695