MHGanainy/mgpt-lora-multi-shared-1024

README.md

@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 # mgpt-lora-multi-shared-1024
 
 This model is a fine-tuned version of [ai-forever/mGPT](https://huggingface.co/ai-forever/mGPT) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.7964
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 3.7963714599609375,
+    "eval_runtime": 10.5803,
+    "eval_samples_per_second": 9.451,
+    "eval_steps_per_second": 9.451,
+    "perplexity": 44.53927837463248,
+    "total_flos": 1527651080601600.0,
+    "train_loss": 4.171203002929688,
+    "train_runtime": 157.8425,
+    "train_samples_per_second": 0.634,
+    "train_steps_per_second": 0.634
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "eval_loss": 3.7963714599609375,
+    "eval_runtime": 10.5803,
+    "eval_samples_per_second": 9.451,
+    "eval_steps_per_second": 9.451,
+    "perplexity": 44.53927837463248
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 1527651080601600.0,
+    "train_loss": 4.171203002929688,
+    "train_runtime": 157.8425,
+    "train_samples_per_second": 0.634,
+    "train_steps_per_second": 0.634
+}

trainer_state.json

@@ -0,0 +1,49 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.4095587730407715,
+      "learning_rate": 0.0,
+      "loss": 4.1712,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "step": 100,
+      "total_flos": 1527651080601600.0,
+      "train_loss": 4.171203002929688,
+      "train_runtime": 157.8425,
+      "train_samples_per_second": 0.634,
+      "train_steps_per_second": 0.634
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1527651080601600.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}