t5-base-spanish-yoremnokki

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: transformers
 license: apache-2.0
-base_model: google-t5/t5-small
+base_model: google-t5/t5-base
 tags:
 - generated_from_trainer
 metrics:
@@ -16,11 +16,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5-small-spanish-yoremnokki
 
-This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an unknown dataset.
+This model is a fine-tuned version of [google-t5/t5-base](https://huggingface.co/google-t5/t5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8387
-- Bleu: 0.0357
-- Gen Len: 11.8931
+- Loss: 1.8622
+- Bleu: 10.7378
+- Gen Len: 14.22
 
 ## Model description
 
@@ -39,7 +39,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
+- learning_rate: 1e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
@@ -47,17 +47,18 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 4
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Bleu   | Gen Len |
-|:-------------:|:------:|:----:|:---------------:|:------:|:-------:|
-| No log        | 0.9989 | 463  | 3.1886          | 0.0016 | 11.7323 |
-| 3.8008        | 2.0    | 927  | 2.9507          | 0.0162 | 11.6761 |
-| 3.2701        | 2.9989 | 1390 | 2.8636          | 0.0301 | 11.8016 |
-| 3.125         | 3.9957 | 1852 | 2.8387          | 0.0357 | 11.8931 |
+| Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
+| 3.4879        | 1.0   | 839  | 2.3617          | 0.2661  | 14.6218 |
+| 2.4168        | 2.0   | 1678 | 2.0854          | 1.8065  | 14.3043 |
+| 2.2109        | 3.0   | 2517 | 1.9512          | 5.2629  | 14.2827 |
+| 2.1335        | 4.0   | 3356 | 1.8816          | 9.4622  | 14.23   |
+| 2.0758        | 5.0   | 4195 | 1.8622          | 10.7378 | 14.22   |
 
 
 ### Framework versions

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "google-t5/t5-small",
+  "_name_or_path": "google-t5/t5-base",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
   "classifier_dropout": 0.0,
-  "d_ff": 2048,
+  "d_ff": 3072,
   "d_kv": 64,
-  "d_model": 512,
+  "d_model": 768,
   "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": 6,
-  "num_heads": 8,
-  "num_layers": 6,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
   "output_past": true,
   "pad_token_id": 0,
   "relative_attention_max_distance": 128,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee16eaeccd1e8db97cf19daf80b6e517b364ea890156e605098c57ab8e10cedd
-size 242041896
+oid sha256:64f25b5668c1a82012682bd2bb466412ae03e4ef20a75ca327ef24a182db0ae9
+size 891644712

runs/Oct04_03-28-34_6b1117a5b990/events.out.tfevents.1728012517.6b1117a5b990.298.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96e926d1204caa71ccc8b4fe354a25c47a1e930f39c2e9a32e92cfa25e49c276
+size 9821

special_tokens_map.json

@@ -101,25 +101,7 @@
     "<extra_id_98>",
     "<extra_id_99>"
   ],
-  "eos_token": {
-    "content": "</s>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "pad_token": {
-    "content": "<pad>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "unk_token": {
-    "content": "<unk>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -930,7 +930,7 @@
   "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "extra_ids": 100,
-  "model_max_length": 512,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "tokenizer_class": "T5Tokenizer",
   "unk_token": "<unk>"

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7544803a830e998e0544574258736b129bd27bff951f70a60b1ffb18efd9a1ca
+oid sha256:0bf69dc795cf26a94c7f4ba4f71d3d7c4ccc11b4e9d85f6612d5fe61a3d358ab
 size 5368