update model

config.json

@@ -1,12 +1,13 @@
 {
   "_name_or_path": "google/long-t5-tglobal-base",
   "architectures": [
-    "T5ForConditionalGeneration"
+    "LongT5ForConditionalGeneration"
   ],
   "d_ff": 2048,
   "d_kv": 64,
   "d_model": 768,
   "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
   "dropout_rate": 0.1,
   "encoder_attention_type": "transient-global",
   "eos_token_id": 1,
@@ -14,9 +15,10 @@
   "global_block_size": 16,
   "initializer_factor": 1.0,
   "is_encoder_decoder": true,
+  "is_gated_act": true,
   "layer_norm_epsilon": 1e-06,
   "local_radius": 127,
-  "model_type": "t5",
+  "model_type": "longt5",
   "n_positions": 4096,
   "num_decoder_layers": 12,
   "num_heads": 12,
@@ -26,8 +28,8 @@
   "relative_attention_max_distance": 128,
   "relative_attention_num_buckets": 32,
   "tie_word_embeddings": false,
-  "torch_dtype": "float16",
-  "transformers_version": "4.18.0",
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.1",
   "use_cache": true,
   "vocab_size": 32128
 }

detail.json

@@ -1,6 +1,6 @@
 {
-    "avg_loss": 1.2271652221679688,
-    "step": 722495,
-    "best_loss": 1.2271652221679688,
+    "avg_loss": 0.15461508115731704,
+    "step": 808767,
+    "best_loss": 0.15461508115731704,
     "lr": 0.0003
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41becfa1f44ef5d8bca54a5cb44e737360e6a4a1abe046efa6c52b5720507142
-size 1187801417
+oid sha256:51d51b601d6bfa225fd117993dfd885b0db8be426eeb35655d52dcd85032af64
+size 990450547