End of training

README.md

@@ -1,13 +1,31 @@
 ---
+language:
+- en
+- he
 license: apache-2.0
 base_model: t5-small
 tags:
 - generated_from_trainer
 datasets:
 - opus100
+metrics:
+- bleu
 model-index:
 - name: english-hebrew
-  results: []
+  results:
+  - task:
+      name: Translation
+      type: translation
+    dataset:
+      name: opus100 en-he
+      type: opus100
+      config: en-he
+      split: validation
+      args: en-he
+    metrics:
+    - name: Bleu
+      type: bleu
+      value: 38.03154544811525
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,7 +33,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # english-hebrew
 
-This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the opus100 dataset.
+This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the opus100 en-he dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2399
+- Bleu: 38.0315
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 3.0,
-    "eval_bleu": 0.0,
-    "eval_loss": 0.36301419138908386,
-    "eval_runtime": 93.3757,
-    "eval_samples": 8,
-    "eval_samples_per_second": 0.086,
-    "eval_steps_per_second": 0.021,
-    "train_loss": 0.2741566179393298,
-    "train_runtime": 830.8843,
-    "train_samples": 3907,
-    "train_samples_per_second": 14.107,
-    "train_steps_per_second": 3.528
+    "eval_bleu": 38.03154544811525,
+    "eval_loss": 0.23992294073104858,
+    "eval_runtime": 138.9929,
+    "eval_samples": 32,
+    "eval_samples_per_second": 0.23,
+    "eval_steps_per_second": 0.058,
+    "train_loss": 0.24653975453639984,
+    "train_runtime": 3373.715,
+    "train_samples": 15625,
+    "train_samples_per_second": 13.894,
+    "train_steps_per_second": 3.474
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 3.0,
-    "eval_bleu": 0.0,
-    "eval_loss": 0.36301419138908386,
-    "eval_runtime": 93.3757,
-    "eval_samples": 8,
-    "eval_samples_per_second": 0.086,
-    "eval_steps_per_second": 0.021
+    "eval_bleu": 38.03154544811525,
+    "eval_loss": 0.23992294073104858,
+    "eval_runtime": 138.9929,
+    "eval_samples": 32,
+    "eval_samples_per_second": 0.23,
+    "eval_steps_per_second": 0.058
 }

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
   "pad_token_id": 0,

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.2741566179393298,
-    "train_runtime": 830.8843,
-    "train_samples": 3907,
-    "train_samples_per_second": 14.107,
-    "train_steps_per_second": 3.528
+    "train_loss": 0.24653975453639984,
+    "train_runtime": 3373.715,
+    "train_samples": 15625,
+    "train_samples_per_second": 13.894,
+    "train_steps_per_second": 3.474
 }

trainer_state.json

@@ -3,56 +3,164 @@
   "best_model_checkpoint": null,
   "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 2931,
+  "global_step": 11721,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.51,
-      "learning_rate": 4.14704878880928e-05,
-      "loss": 0.3586,
+      "epoch": 0.13,
+      "learning_rate": 4.7867076188038565e-05,
+      "loss": 0.3426,
       "step": 500
     },
     {
-      "epoch": 1.02,
-      "learning_rate": 3.294097577618561e-05,
-      "loss": 0.2728,
+      "epoch": 0.26,
+      "learning_rate": 4.573415237607713e-05,
+      "loss": 0.2889,
       "step": 1000
     },
     {
-      "epoch": 1.54,
-      "learning_rate": 2.4411463664278406e-05,
-      "loss": 0.2668,
+      "epoch": 0.38,
+      "learning_rate": 4.360122856411569e-05,
+      "loss": 0.2868,
       "step": 1500
     },
     {
-      "epoch": 2.05,
-      "learning_rate": 1.5881951552371206e-05,
+      "epoch": 0.51,
+      "learning_rate": 4.146830475215426e-05,
       "loss": 0.2508,
       "step": 2000
     },
     {
-      "epoch": 2.56,
-      "learning_rate": 7.3524394404640055e-06,
-      "loss": 0.2471,
+      "epoch": 0.64,
+      "learning_rate": 3.933538094019282e-05,
+      "loss": 0.2525,
       "step": 2500
     },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.720245712823138e-05,
+      "loss": 0.2547,
+      "step": 3000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.506953331626994e-05,
+      "loss": 0.2494,
+      "step": 3500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2936609504308506e-05,
+      "loss": 0.2532,
+      "step": 4000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 3.0803685692347076e-05,
+      "loss": 0.2496,
+      "step": 4500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.8670761880385632e-05,
+      "loss": 0.2321,
+      "step": 5000
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 2.6537838068424198e-05,
+      "loss": 0.244,
+      "step": 5500
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 2.440491425646276e-05,
+      "loss": 0.2433,
+      "step": 6000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 2.2271990444501324e-05,
+      "loss": 0.237,
+      "step": 6500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.0139066632539887e-05,
+      "loss": 0.2308,
+      "step": 7000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.800614282057845e-05,
+      "loss": 0.225,
+      "step": 7500
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 1.5873219008617013e-05,
+      "loss": 0.2378,
+      "step": 8000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 1.3740295196655576e-05,
+      "loss": 0.2264,
+      "step": 8500
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.1607371384694139e-05,
+      "loss": 0.2289,
+      "step": 9000
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 9.474447572732703e-06,
+      "loss": 0.2321,
+      "step": 9500
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 7.3415237607712656e-06,
+      "loss": 0.2356,
+      "step": 10000
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.2085999488098285e-06,
+      "loss": 0.2279,
+      "step": 10500
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 3.075676136848392e-06,
+      "loss": 0.2194,
+      "step": 11000
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 9.42752324886955e-07,
+      "loss": 0.228,
+      "step": 11500
+    },
     {
       "epoch": 3.0,
-      "step": 2931,
-      "total_flos": 74769472487424.0,
-      "train_loss": 0.2741566179393298,
-      "train_runtime": 830.8843,
-      "train_samples_per_second": 14.107,
-      "train_steps_per_second": 3.528
+      "step": 11721,
+      "total_flos": 298502687293440.0,
+      "train_loss": 0.24653975453639984,
+      "train_runtime": 3373.715,
+      "train_samples_per_second": 13.894,
+      "train_steps_per_second": 3.474
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2931,
+  "max_steps": 11721,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 74769472487424.0,
+  "total_flos": 298502687293440.0,
   "trial_name": null,
   "trial_params": null
 }