Upload 7 files

config.json

@@ -1,12 +1,11 @@
 {
-  "_name_or_path": "google/flan-t5-large",
+  "_name_or_path": "google/flan-t5-xl",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
-  "classifier_dropout": 0.0,
-  "d_ff": 2816,
+  "d_ff": 5120,
   "d_kv": 64,
-  "d_model": 1024,
+  "d_model": 2048,
   "decoder_start_token_id": 0,
   "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
@@ -17,18 +16,46 @@
   "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
-  "max_length": 1024,
   "n_positions": 512,
   "num_decoder_layers": 24,
-  "num_heads": 16,
+  "num_heads": 32,
   "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: "
+    }
+  },
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.33.1",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.27.4",
   "use_cache": false,
   "vocab_size": 32128
 }

generation_config.json

@@ -2,7 +2,6 @@
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
   "pad_token_id": 0,
-  "max_length": 1024,
-  "transformers_version": "4.33.1",
+  "transformers_version": "4.27.4",
   "use_cache": false
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0959c3c9892cb661f09de749171b9a7961ca0ff9448f2433bac49c586ca8601
-size 3132793669
+oid sha256:08cb0eec9ab7f9f6e8c063f76224e4c426abb500e65b48efee38825dcbbf45d2
+size 11399218325

tokenizer_config.json

@@ -101,12 +101,12 @@
     "<extra_id_98>",
     "<extra_id_99>"
   ],
-  "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "extra_ids": 100,
   "model_max_length": 512,
   "pad_token": "<pad>",
   "sp_model_kwargs": {},
+  "special_tokens_map_file": "/home/arthur_huggingface_co/.cache/huggingface/hub/models--google--t5-v1_1-small/snapshots/fb7e6cba609f7bab11c614294bc04f82f613c7b1/special_tokens_map.json",
   "tokenizer_class": "T5Tokenizer",
   "unk_token": "<unk>"
 }