Training in progress, step 500

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "t5-base",
+  "_name_or_path": "t5-small",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "d_ff": 3072,
+  "d_ff": 2048,
   "d_kv": 64,
-  "d_model": 768,
+  "d_model": 512,
   "decoder_start_token_id": 0,
   "dense_act_fn": "relu",
   "dropout_rate": 0.1,
@@ -18,9 +18,9 @@
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
   "n_positions": 512,
-  "num_decoder_layers": 12,
-  "num_heads": 12,
-  "num_layers": 12,
+  "num_decoder_layers": 6,
+  "num_heads": 8,
+  "num_layers": 6,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f973a2126952e9043e4dac0d48278cf0f20324ac78ea30331aaca04cd50533cd
-size 891702929
+oid sha256:2a9b7fa9c25193ba8f357d9441156de71f07b773a958e3a4c5171db97ad97af0
+size 242071641

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3173b658398e6810481b9826e591a182541d1f8f83d392fb126555cfa275c75
+oid sha256:757a003888d16bec236b415e80264479d9c6278544316cbb2f66825a49f6fd7a
 size 4219