Training in progress, epoch 1

config.json

@@ -3,9 +3,9 @@
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "d_ff": 512,
-  "d_kv": 16,
-  "d_model": 128,
+  "d_ff": 1024,
+  "d_kv": 64,
+  "d_model": 256,
   "decoder_start_token_id": 259,
   "dense_act_fn": "relu",
   "dropout_rate": 0.3,
@@ -16,9 +16,9 @@
   "is_gated_act": false,
   "layer_norm_epsilon": 1e-06,
   "model_type": "t5",
-  "num_decoder_layers": 6,
-  "num_heads": 6,
-  "num_layers": 6,
+  "num_decoder_layers": 4,
+  "num_heads": 4,
+  "num_layers": 4,
   "pad_token_id": 0,
   "relative_attention_max_distance": 64,
   "relative_attention_num_buckets": 16,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9359a363befd957da7b42b05a48bf15d7ebfc123ecab83e890d068125fcff0ff
-size 10059592
+oid sha256:464ddd05bf59e3211072364dc59ff0cfa4ed1f7dd9bcbe06568bcd5c3fbf4b58
+size 29787736

runs/Nov23_06-52-32_3a71d3232b61/events.out.tfevents.1700722361.3a71d3232b61.1811.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f685c6432689a28c22acf2337e2b50b2ad81dec6bec09caf5fa34c9ec7bb1ee
+size 4780

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6de771aa37db4a1a743cb122fe79bd7ee149348933c60714d2f3b69cde1563bd
+oid sha256:12424ea04b3b636595515d470a771cc36ba4c1a2d1fbfc16913bebc1d97e90d3
 size 4728