update model card README.md

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on the xtreme_s dataset.
 It achieves the following results on the evaluation set:
+- Loss: 1.7768
 - Bleu: 0.0000
-- Loss: 1.6425
 
 ## Model description
 
@@ -37,12 +37,12 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.00023509256443134124
+- learning_rate: 0.0003
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 64
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
@@ -51,27 +51,17 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Bleu   | Validation Loss |
-|:-------------:|:-----:|:----:|:------:|:---------------:|
-| 2.8303        | 0.15  | 500  | 0.0    | 4.9238          |
-| 2.4062        | 0.31  | 1000 | 0.0    | 4.4017          |
-| 1.9171        | 0.46  | 1500 | 0.0000 | 3.6431          |
-| 1.4558        | 0.62  | 2000 | 0.0000 | 2.8292          |
-| 1.2393        | 0.77  | 2500 | 0.0000 | 2.3704          |
-| 1.3315        | 0.93  | 3000 | 0.0000 | 2.3101          |
-| 1.8476        | 1.08  | 3500 | 0.0000 | 1.9936          |
-| 1.683         | 1.23  | 4000 | 0.0000 | 1.9308          |
-| 1.8298        | 1.39  | 4500 | 0.0000 | 1.8817          |
-| 1.5955        | 1.54  | 5000 | 0.0000 | 1.8171          |
-| 1.6288        | 1.7   | 5500 | 0.0000 | 1.7821          |
-| 1.4107        | 1.85  | 6000 | 0.0000 | 1.7170          |
-| 1.0363        | 2.01  | 6500 | 0.0000 | 1.7419          |
-| 0.9667        | 2.16  | 7000 | 0.0000 | 1.7309          |
-| 0.9147        | 2.31  | 7500 | 0.0000 | 1.7244          |
-| 1.1975        | 2.47  | 8000 | 0.0000 | 1.6716          |
-| 0.9071        | 2.62  | 8500 | 0.0000 | 1.6886          |
-| 0.9735        | 2.78  | 9000 | 0.0000 | 1.6609          |
-| 0.908         | 2.93  | 9500 | 0.0000 | 1.6425          |
+| Training Loss | Epoch | Step | Validation Loss | Bleu   |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 2.5511        | 0.31  | 500  | 5.1039          | 0.0    |
+| 2.2033        | 0.62  | 1000 | 4.1782          | 0.0000 |
+| 1.4703        | 0.93  | 1500 | 2.8979          | 0.0000 |
+| 1.6507        | 1.23  | 2000 | 2.2250          | 0.0000 |
+| 1.6791        | 1.54  | 2500 | 2.0530          | 0.0000 |
+| 1.4587        | 1.85  | 3000 | 1.9121          | 0.0000 |
+| 1.288         | 2.16  | 3500 | 1.8705          | 0.0000 |
+| 1.2244        | 2.47  | 4000 | 1.7940          | 0.0000 |
+| 1.0364        | 2.78  | 4500 | 1.7768          | 0.0000 |
 
 
 ### Framework versions

wandb/run-20220505_173818-i9acyhfo/files/output.log

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2192ab1040218726cf08d6108176c5547d818ea56f5f7aa9b6813a899189b0b8
-size 16288756
+oid sha256:e89684c4b6746ae738e3ddaa427cf9d90c966dd964975e6c83cc9286c986321c
+size 16290323

wandb/run-20220505_173818-i9acyhfo/logs/debug-internal.log

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e883f57d9e77232f2cf3a3fde646766c69e97db702e1b1a7c7d12e4b494f669e
-size 14780150
+oid sha256:51104230cbff36190fb046f99c188978d0072b182cec0fb29ce099113c8c84d1
+size 14783102

wandb/run-20220505_173818-i9acyhfo/run-i9acyhfo.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:654fb5c5647c98ae01af1ccbafa2354d65d474f4c254d33a30c76dcf57ab6ee4
-size 1045114644
+oid sha256:19b65ff5e38c62ffeab088591c02c2726ee560f2f55bbf6e994f8ca1036f5c6a
+size 1045118132