huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/rihanna")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1yq53sqb/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3g4a2qa9/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
 
 ## Training procedure
 
 The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Rihanna's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2yzmvwft) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3du2fhxz) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2yzmvwft/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3du2fhxz/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/rihanna",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.336484432220459, "eval_runtime": 8.2149, "eval_samples_per_second": 22.155, "eval_steps_per_second": 2.8, "epoch": 1.0}
+{"eval_loss": 2.237405776977539, "eval_runtime": 8.4305, "eval_samples_per_second": 22.181, "eval_steps_per_second": 2.847, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09d9a3dd92329367c13eceb5662222bc53e8696915c8a5f7c415b644ae8db1a5
+oid sha256:fd6be841db360da1a00ad23a76226c5b19e42247bc72575b241a6f446840b95d
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe58cc1fdcfd3ac06e70b1529c91aa020d938437f9d3b29e464ebded997e5c43
-size 995603825
+oid sha256:7a71909dbe3a3827eb2d96903ec0cd4f6475fbd143d3adcebee21e68eb480324
+size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a08b47a46d68a78c664567394101b6738c8c5b9b75221e0f88b43153005280f
+oid sha256:c5c96ee233c3281090b842a4a11ff662f1751f775bd8bd50db7037dbc4bdf1d8
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d50bc04cec0dd1eaca41f1aa92a674115be7156b55479eca90b612cd63c8c333
+oid sha256:26ebd8f09a6c1db84aa3aaeb1ea1345bcb3c2ae6c124b74ff19c3f608acd9e3c
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:874f414e996bed5c01ac5b2a3b90399694a3bb1b56ce1bf829358b7185235dd2
+oid sha256:0b2179c852b9a3c2fc23aeed095aeb8cf7d63d1da7d9c6eca4435ee27bc43703
 size 623

tokenizer_config.json

@@ -1 +1 @@
-{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/rihanna", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.336484432220459,
-  "best_model_checkpoint": "output/rihanna/checkpoint-131",
-  "epoch": 1.0,
-  "global_step": 131,
+  "best_metric": 2.237405776977539,
+  "best_model_checkpoint": "output/rihanna/checkpoint-262",
+  "epoch": 2.0,
+  "global_step": 262,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -170,11 +170,175 @@
       "eval_samples_per_second": 22.593,
       "eval_steps_per_second": 2.855,
       "step": 131
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.153829445781081e-07,
+      "loss": 2.3816,
+      "step": 135
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.5916577414195624e-06,
+      "loss": 2.1418,
+      "step": 140
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.8302217539407e-06,
+      "loss": 2.1866,
+      "step": 145
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 6.998927551907465e-06,
+      "loss": 2.3514,
+      "step": 150
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.1052270183036815e-05,
+      "loss": 2.3302,
+      "step": 155
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.5932040657672757e-05,
+      "loss": 2.0237,
+      "step": 160
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.1568161872809022e-05,
+      "loss": 2.1338,
+      "step": 165
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.7879694970972374e-05,
+      "loss": 2.1066,
+      "step": 170
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.477600168191081e-05,
+      "loss": 2.1939,
+      "step": 175
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.215804595500327e-05,
+      "loss": 2.1481,
+      "step": 180
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.991981618998877e-05,
+      "loss": 2.4364,
+      "step": 185
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.794984764173325e-05,
+      "loss": 2.1677,
+      "step": 190
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 6.613282313617852e-05,
+      "loss": 2.3095,
+      "step": 195
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 7.435122911001204e-05,
+      "loss": 2.4404,
+      "step": 200
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 8.248704319210595e-05,
+      "loss": 2.0718,
+      "step": 205
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.042342909181217e-05,
+      "loss": 2.2015,
+      "step": 210
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.804641445426643e-05,
+      "loss": 2.1899,
+      "step": 215
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00010524652758746261,
+      "loss": 2.1465,
+      "step": 220
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00011192036955648884,
+      "loss": 2.3478,
+      "step": 225
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00011797209906849287,
+      "loss": 2.3704,
+      "step": 230
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.0001233148088243337,
+      "loss": 2.2886,
+      "step": 235
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00012787177357149405,
+      "loss": 2.1345,
+      "step": 240
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.00013157755193529395,
+      "loss": 2.0747,
+      "step": 245
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00013437892620529645,
+      "loss": 2.2118,
+      "step": 250
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.000136235666580879,
+      "loss": 2.1172,
+      "step": 255
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.00013712110890084145,
+      "loss": 2.2353,
+      "step": 260
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.237405776977539,
+      "eval_runtime": 8.3034,
+      "eval_samples_per_second": 22.521,
+      "eval_steps_per_second": 2.89,
+      "step": 262
     }
   ],
-  "max_steps": 131,
-  "num_train_epochs": 1,
-  "total_flos": 136786378752000.0,
+  "max_steps": 262,
+  "num_train_epochs": 2,
+  "total_flos": 272919527424000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7cde27565a440c8641232e5c6caa577dc184d1b7faac3fc46fa7661408a4403
+oid sha256:74c710dd5a8d4b67cb50ad0843ddfa8eb650173d3491cfb493f1dda199d69582
 size 2671