End of training

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.0285
+- Loss: 0.1668
 
 ## Model description
 
@@ -39,12 +39,19 @@ The following hyperparameters were used during training:
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 128
+- total_eval_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 0.001
+- num_epochs: 10.0
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.4073        | 6.33  | 500  | 0.1755          |
 
 
 ### Framework versions

adapter_config.json

@@ -19,8 +19,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q",
-    "v"
+    "v",
+    "q"
   ],
   "task_type": "SEQ_2_SEQ_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ed3301e7ce1a78f37bd682a8fecf5e91b951e21ea2a9b2d43f49e3bc69baa77
+oid sha256:b665b8f584f09c5ba48c254e7814251fa797afeb904c0b34a1402cd9362d057f
 size 2765880

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 0.0,
-    "eval_loss": 2.0285420417785645,
-    "eval_runtime": 13.2838,
-    "eval_samples_per_second": 591.473,
-    "eval_steps_per_second": 37.038,
-    "train_loss": 2.6510977745056152,
-    "train_runtime": 0.8011,
-    "train_samples_per_second": 12.483,
-    "train_steps_per_second": 1.248
+    "epoch": 10.0,
+    "eval_loss": 0.1668280065059662,
+    "eval_runtime": 1.7776,
+    "eval_samples_per_second": 4420.127,
+    "eval_steps_per_second": 34.879,
+    "train_loss": 0.35002156511137755,
+    "train_runtime": 70.584,
+    "train_samples_per_second": 1416.752,
+    "train_steps_per_second": 11.192
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.0,
-    "eval_loss": 2.0285420417785645,
-    "eval_runtime": 13.2838,
-    "eval_samples_per_second": 591.473,
-    "eval_steps_per_second": 37.038
+    "epoch": 10.0,
+    "eval_loss": 0.1668280065059662,
+    "eval_runtime": 1.7776,
+    "eval_samples_per_second": 4420.127,
+    "eval_steps_per_second": 34.879
 }

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.0,
-    "train_loss": 2.6510977745056152,
-    "train_runtime": 0.8011,
-    "train_samples_per_second": 12.483,
-    "train_steps_per_second": 1.248
+    "epoch": 10.0,
+    "train_loss": 0.35002156511137755,
+    "train_runtime": 70.584,
+    "train_samples_per_second": 1416.752,
+    "train_steps_per_second": 11.192
 }

trainer_state.json

@@ -1,29 +1,44 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0016,
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 1,
+  "global_step": 790,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0,
-      "step": 1,
-      "total_flos": 517012193280.0,
-      "train_loss": 2.6510977745056152,
-      "train_runtime": 0.8011,
-      "train_samples_per_second": 12.483,
-      "train_steps_per_second": 1.248
+      "epoch": 6.33,
+      "grad_norm": 0.20978006720542908,
+      "learning_rate": 0.0003670886075949367,
+      "loss": 0.4073,
+      "step": 500
+    },
+    {
+      "epoch": 6.33,
+      "eval_loss": 0.1755284070968628,
+      "eval_runtime": 1.8212,
+      "eval_samples_per_second": 4314.295,
+      "eval_steps_per_second": 34.044,
+      "step": 500
+    },
+    {
+      "epoch": 10.0,
+      "step": 790,
+      "total_flos": 3267517032169472.0,
+      "train_loss": 0.35002156511137755,
+      "train_runtime": 70.584,
+      "train_samples_per_second": 1416.752,
+      "train_steps_per_second": 11.192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1,
+  "max_steps": 790,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 517012193280.0,
+  "total_flos": 3267517032169472.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acdf1673b45bf8cf82f8d15db1b4588093fb6a9c8de2bf726eb15278ebb70f95
+oid sha256:50056c87829ee7b1ccaaf40166a7f45dc67bf22c7549e07f6ec646d078d3eaaf
 size 5112