End of training

README.md

@@ -14,7 +14,7 @@ model-index:
       name: Causal Language Modeling
       type: text-generation
     dataset:
-      name: wikitext
+      name: wikitext wikitext-2-raw-v1
       type: wikitext
       config: wikitext-2-raw-v1
       split: validation
@@ -30,7 +30,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # clm
 
-This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the wikitext dataset.
+This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the wikitext wikitext-2-raw-v1 dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.4326
 - Accuracy: 0.3767

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.3767106549364614,
+    "eval_loss": 3.4325666427612305,
+    "eval_runtime": 11.7361,
+    "eval_samples": 240,
+    "eval_samples_per_second": 20.45,
+    "eval_steps_per_second": 2.556,
+    "perplexity": 30.955993850684628,
+    "total_flos": 605685867872256.0,
+    "train_loss": 3.6505329921327787,
+    "train_runtime": 308.0636,
+    "train_samples": 2318,
+    "train_samples_per_second": 7.524,
+    "train_steps_per_second": 0.471
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.3767106549364614,
+    "eval_loss": 3.4325666427612305,
+    "eval_runtime": 11.7361,
+    "eval_samples": 240,
+    "eval_samples_per_second": 20.45,
+    "eval_steps_per_second": 2.556,
+    "perplexity": 30.955993850684628
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 605685867872256.0,
+    "train_loss": 3.6505329921327787,
+    "train_runtime": 308.0636,
+    "train_samples": 2318,
+    "train_samples_per_second": 7.524,
+    "train_steps_per_second": 0.471
+}

trainer_state.json

@@ -0,0 +1,121 @@
+{
+  "best_metric": 0.3767106549364614,
+  "best_model_checkpoint": "D:/1_SyscoPY_D/NLP/Data/Transformers_Hug/checkpoint/distilgpt2\\checkpoint-145",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 145,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 3.9461,
+      "step": 10
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 2.884615384615385e-05,
+      "loss": 3.813,
+      "step": 20
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.6538461538461538e-05,
+      "loss": 3.6952,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 2.423076923076923e-05,
+      "loss": 3.6721,
+      "step": 40
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.1923076923076924e-05,
+      "loss": 3.6339,
+      "step": 50
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9615384615384617e-05,
+      "loss": 3.6567,
+      "step": 60
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7307692307692306e-05,
+      "loss": 3.6063,
+      "step": 70
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.5e-05,
+      "loss": 3.6228,
+      "step": 80
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.2692307692307693e-05,
+      "loss": 3.5844,
+      "step": 90
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.0384615384615384e-05,
+      "loss": 3.6194,
+      "step": 100
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 8.076923076923077e-06,
+      "loss": 3.5918,
+      "step": 110
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 5.76923076923077e-06,
+      "loss": 3.5682,
+      "step": 120
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.4615384615384617e-06,
+      "loss": 3.5482,
+      "step": 130
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 3.585,
+      "step": 140
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3767106549364614,
+      "eval_loss": 3.4325666427612305,
+      "eval_runtime": 12.5136,
+      "eval_samples_per_second": 19.179,
+      "eval_steps_per_second": 2.397,
+      "step": 145
+    },
+    {
+      "epoch": 1.0,
+      "step": 145,
+      "total_flos": 605685867872256.0,
+      "train_loss": 3.6505329921327787,
+      "train_runtime": 308.0636,
+      "train_samples_per_second": 7.524,
+      "train_steps_per_second": 0.471
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 145,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 605685867872256.0,
+  "trial_name": null,
+  "trial_params": null
+}