Model update

New model files after re-training.

README.md

@@ -4,6 +4,52 @@ language:
 widget:
 - text: "A:The rest of it is just like wild flowers\nB:Wild flowers.\nA:and in the summertime we get the prettiest yellow flowers.\nB:Uh-huh.\nA:They're about, oh, I guess, a third the size of your, palm of your hand\nB:"
   example_title: "SW Example 1"
+metrics:
+- accuracy
+model-index:
+- name: gpt2-small-sw
+  results: []
 ---
 
-GPT-2 small fine-tuned on SwitchBoard. More details soon.
+# gpt2-small-sw
+
+This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the SwitchBoard dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.4870
+- Accuracy: 0.4657
+
+## Model description
+
+More details soon.
+
+## Intended uses & limitations
+
+More details soon.
+
+## Training and evaluation data
+
+More details soon.
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 5
+- eval_batch_size: 5
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 20.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.26.0
+- Pytorch 1.13.1
+- Datasets 2.9.0
+- Tokenizers 0.13.0.dev0

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.46566471163245354,
+    "eval_loss": 2.4869625568389893,
+    "eval_runtime": 13.4779,
+    "eval_samples": 208,
+    "eval_samples_per_second": 15.433,
+    "eval_steps_per_second": 3.116,
+    "perplexity": 12.024696262523243,
+    "train_loss": 2.1814393838702095,
+    "train_runtime": 6610.5408,
+    "train_samples": 1741,
+    "train_samples_per_second": 5.267,
+    "train_steps_per_second": 1.056
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.46566471163245354,
+    "eval_loss": 2.4869625568389893,
+    "eval_runtime": 13.4779,
+    "eval_samples": 208,
+    "eval_samples_per_second": 15.433,
+    "eval_steps_per_second": 3.116,
+    "perplexity": 12.024696262523243
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99172dbbb8273ca8b7519694ccec9b74826c9f3d1fe6130ac6d100ee158d1d1f
+oid sha256:4c3f088a4fd9ff70b6ed32c3386721c6e819adb994587e1d2901927f01441cce
 size 510398013

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 2.1814393838702095,
+    "train_runtime": 6610.5408,
+    "train_samples": 1741,
+    "train_samples_per_second": 5.267,
+    "train_steps_per_second": 1.056
+}

trainer_state.json

@@ -0,0 +1,103 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "global_step": 6980,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.43,
+      "learning_rate": 4.641833810888253e-05,
+      "loss": 2.5428,
+      "step": 500
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 4.2836676217765046e-05,
+      "loss": 2.4266,
+      "step": 1000
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.9255014326647564e-05,
+      "loss": 2.3389,
+      "step": 1500
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 3.567335243553009e-05,
+      "loss": 2.2819,
+      "step": 2000
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 3.2091690544412614e-05,
+      "loss": 2.2307,
+      "step": 2500
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 2.851002865329513e-05,
+      "loss": 2.1831,
+      "step": 3000
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 2.492836676217765e-05,
+      "loss": 2.1536,
+      "step": 3500
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 2.1346704871060173e-05,
+      "loss": 2.112,
+      "step": 4000
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.7765042979942695e-05,
+      "loss": 2.0913,
+      "step": 4500
+    },
+    {
+      "epoch": 14.33,
+      "learning_rate": 1.4183381088825215e-05,
+      "loss": 2.0638,
+      "step": 5000
+    },
+    {
+      "epoch": 15.76,
+      "learning_rate": 1.0601719197707736e-05,
+      "loss": 2.0455,
+      "step": 5500
+    },
+    {
+      "epoch": 17.19,
+      "learning_rate": 7.020057306590258e-06,
+      "loss": 2.0315,
+      "step": 6000
+    },
+    {
+      "epoch": 18.62,
+      "learning_rate": 3.4383954154727795e-06,
+      "loss": 2.0204,
+      "step": 6500
+    },
+    {
+      "epoch": 20.0,
+      "step": 6980,
+      "total_flos": 1.819637710848e+16,
+      "train_loss": 2.1814393838702095,
+      "train_runtime": 6610.5408,
+      "train_samples_per_second": 5.267,
+      "train_steps_per_second": 1.056
+    }
+  ],
+  "max_steps": 6980,
+  "num_train_epochs": 20,
+  "total_flos": 1.819637710848e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:026aad8124735823a666711e06652a0cb17a71b0c859e2eaf24c07340a38421d
-size 3579
+oid sha256:93b93c6a6f2dbe5baaf363ffad1e01f21c623144709e505eeec0207b9978732e
+size 3515