Training in progress, step 400

config.json

@@ -3,9 +3,9 @@
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "d_ff": 256,
+  "d_ff": 1024,
   "d_kv": 64,
-  "d_model": 32,
+  "d_model": 256,
   "decoder_start_token_id": 55,
   "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": 2,
-  "num_heads": 2,
-  "num_layers": 2,
+  "num_decoder_layers": 4,
+  "num_heads": 4,
+  "num_layers": 4,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,
   "relative_attention_num_buckets": 32,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5102792e883d5acfe335d519ae51a513811effd16b8f7759623f405968de6cd3
-size 670200
+oid sha256:1a42de00ceb02250f18280babab04d47636a91dbff283d1a1e4b6710c5412d5d
+size 29454424

runs/Nov28_12-30-09_d7c5ae20caf1/events.out.tfevents.1701174618.d7c5ae20caf1.3157.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3771f0759d4cddab4b306df8ceb17fba3ea5f1d04535f99615c8123071016863
+size 4791

special_tokens_map.json

@@ -1,30 +1,30 @@
 {
   "bos_token": {
     "content": "<s>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": true,
     "single_word": false
   },
   "eos_token": {
     "content": "</s>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": true,
     "single_word": false
   },
   "pad_token": {
     "content": "<pad>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": true,
     "single_word": false
   },
   "unk_token": {
     "content": "<unk>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": true,
     "single_word": false
   }
 }

tokenizer_config.json

@@ -2,33 +2,33 @@
   "added_tokens_decoder": {
     "55": {
       "content": "<s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
     "56": {
       "content": "</s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
     "57": {
       "content": "<unk>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
     "58": {
       "content": "<pad>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
       "single_word": false,
       "special": true
     }
@@ -36,9 +36,10 @@
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
+  "extra_ids": 0,
   "max_len": 128,
   "model_max_length": 128,
   "pad_token": "<pad>",
-  "tokenizer_class": "FixedVocabTokenizer",
+  "tokenizer_class": "CustomTokenizer",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:629b4114042edbf424ed66b9cafd910a1b227379ef25384e2b86246ccc3fa275
-size 4728
+oid sha256:d5c85c88665e8405b7ec073e5f6c58b51d66afe10e3ed8bf4f7f3cf841348dd8
+size 4792

vocab.json

@@ -1 +1 @@
-{"F": 0, "G": 1, ";": 2, "\u00ed": 3, "y": 4, " ": 5, "V": 6, "L": 7, "j": 8, "N": 9, "\u00e9": 10, "l": 11, "n": 12, "c": 13, "M": 14, "3": 15, "4": 16, "I": 17, "\u0142": 18, "w": 19, "\u01eb": 20, "a": 21, "s": 22, "R": 23, "\u0301": 24, "m": 25, "z": 26, "O": 27, "\u012f": 28, "(": 29, "d": 30, "h": 31, "b": 32, "\u0105": 33, ",": 34, "P": 35, ")": 36, "o": 37, "t": 38, "T": 39, "2": 40, "\u00f3": 41, "k": 42, "E": 43, "\u02bc": 44, "S": 45, "e": 46, "\u00e1": 47, "\u0144": 48, "g": 49, "x": 50, "D": 51, "\u0119": 52, "i": 53, "1": 54, "<s>": 55, "</s>": 56, "<unk>": 57, "<pad>": 58}
+{"w": 0, "l": 1, ",": 2, "x": 3, "\u00ed": 4, "d": 5, "\u00e9": 6, "g": 7, "\u01eb": 8, "T": 9, "N": 10, "s": 11, ")": 12, "I": 13, "j": 14, "\u0142": 15, "\u0144": 16, " ": 17, "E": 18, "\u00e1": 19, "L": 20, "D": 21, "z": 22, "M": 23, "y": 24, "k": 25, "a": 26, "1": 27, "i": 28, "P": 29, "\u012f": 30, "S": 31, "3": 32, "4": 33, "c": 34, "V": 35, "\u0105": 36, "t": 37, "b": 38, "e": 39, "\u02bc": 40, "n": 41, "O": 42, "\u0301": 43, "h": 44, "m": 45, "R": 46, "G": 47, ";": 48, "(": 49, "2": 50, "F": 51, "\u0119": 52, "\u00f3": 53, "o": 54, "<s>": 55, "</s>": 56, "<unk>": 57, "<pad>": 58}