huggingartists

README.md

@@ -14,7 +14,7 @@ widget:
 <div class="inline-flex flex-col" style="line-height: 1.5;">
     <div class="flex">
         <div
-			style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/3c1f124fcbbc2857a95e513fb34cc5a8.400x400x1.jpg&#39;)">
+			style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/721a6c465a666419bf286b473287c33f.446x446x1.jpg&#39;)">
         </div>
     </div>
     <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/taylor-swift")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/orx9pgcj/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/11hietbj/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 Taylor Swift's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/13d0g1o4) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/jwz5zda0) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/13d0g1o4/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/jwz5zda0/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "taylor-swift",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -36,7 +36,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.12.2",
+  "transformers_version": "4.16.2",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.1195662021636963, "eval_runtime": 2.7951, "eval_samples_per_second": 76.561, "eval_steps_per_second": 9.66, "epoch": 5.0}
+{"eval_loss": 1.4377235174179077, "eval_runtime": 9.5718, "eval_samples_per_second": 20.895, "eval_steps_per_second": 2.612, "epoch": 7.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:283ca8f9344b855954167e60f2db67ac2afc241037fe870abcdcea59b7262caf
+oid sha256:76ae138578ec229ab199e35e58682cc0818297990880e98b41bb7caca21799bb
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:642b9dfcc4d3d148df04dbd5dd9fc2e06a625d5877f48239cd186fadbd82f0c0
+oid sha256:fd96b2ecf6961b37a51c11488af2110f05d27248a9d735d9cec37797c7c45cbe
 size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:875d52e81b7c5b242a7496cd7be0fe29bfb0475bf80207c43583aeaede1cdaed
+oid sha256:a05cf58e9ce70f6bc6054d004220f5b59634b254e9d1c7c20c77dd1d160dacdc
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8108166cfcb949d51c2438ab25802532e6d155d969ce33575bd2e1f5af1b15a0
+oid sha256:0917e5420fc90d80d96ba9647582ee18ea3c8fd35cfb2e8cff174e79d4c678c0
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7733e18eff27c9a4da0b58764924ec541924a4be41ac9a645f4ee09be1782f09
+oid sha256:696255b08bfcc0a618196cb4d29aaed44996ae47e5ca36d14f60bf60ece9f170
 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/taylor-swift", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.1195662021636963,
-  "best_model_checkpoint": "output/taylor-swift/checkpoint-790",
-  "epoch": 5.0,
-  "global_step": 790,
+  "best_metric": 1.4377235174179077,
+  "best_model_checkpoint": "output/taylor-swift/checkpoint-1120",
+  "epoch": 7.0,
+  "global_step": 1120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -994,11 +994,431 @@
       "eval_samples_per_second": 76.91,
       "eval_steps_per_second": 9.704,
       "step": 790
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 3.303277502872983e-07,
+      "loss": 1.9264,
+      "step": 795
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 1.9314,
+      "step": 800
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.466734766960144,
+      "eval_runtime": 9.3674,
+      "eval_samples_per_second": 21.351,
+      "eval_steps_per_second": 2.669,
+      "step": 800
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 3.303277502872907e-07,
+      "loss": 1.908,
+      "step": 805
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 1.3181297643383773e-06,
+      "loss": 1.8545,
+      "step": 810
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 2.9538929687704367e-06,
+      "loss": 2.0338,
+      "step": 815
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 5.221864069725677e-06,
+      "loss": 1.7718,
+      "step": 820
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 8.100201265702767e-06,
+      "loss": 1.8307,
+      "step": 825
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 1.1561184596045435e-05,
+      "loss": 1.8088,
+      "step": 830
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 1.5571482899316272e-05,
+      "loss": 1.6236,
+      "step": 835
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.0092474810602853e-05,
+      "loss": 2.1289,
+      "step": 840
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.5080620706373914e-05,
+      "loss": 1.4544,
+      "step": 845
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 3.048788201485526e-05,
+      "loss": 2.0407,
+      "step": 850
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 3.6262183853736515e-05,
+      "loss": 1.8147,
+      "step": 855
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 4.2347916539754777e-05,
+      "loss": 1.8843,
+      "step": 860
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 4.868647114034379e-05,
+      "loss": 1.8188,
+      "step": 865
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 5.521680390969348e-05,
+      "loss": 1.8368,
+      "step": 870
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 6.1876024173392e-05,
+      "loss": 1.8814,
+      "step": 875
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 6.859999999999984e-05,
+      "loss": 1.9456,
+      "step": 880
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 7.532397582660791e-05,
+      "loss": 1.6534,
+      "step": 885
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 8.198319609030643e-05,
+      "loss": 1.8587,
+      "step": 890
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 8.851352885965613e-05,
+      "loss": 2.0249,
+      "step": 895
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 9.485208346024515e-05,
+      "loss": 1.9597,
+      "step": 900
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 0.00010093781614626339,
+      "loss": 1.499,
+      "step": 905
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 0.00010671211798514466,
+      "loss": 1.6634,
+      "step": 910
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 0.00011211937929362601,
+      "loss": 1.789,
+      "step": 915
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 0.00011710752518939709,
+      "loss": 1.6422,
+      "step": 920
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 0.00012162851710068368,
+      "loss": 1.83,
+      "step": 925
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 0.00012563881540395453,
+      "loss": 1.8398,
+      "step": 930
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 0.0001290997987342972,
+      "loss": 1.8865,
+      "step": 935
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 0.0001319781359302743,
+      "loss": 2.0384,
+      "step": 940
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 0.00013424610703122953,
+      "loss": 1.8054,
+      "step": 945
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 0.0001358818702356616,
+      "loss": 1.6484,
+      "step": 950
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 0.0001368696722497127,
+      "loss": 2.0007,
+      "step": 955
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.8572,
+      "step": 960
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.5031535625457764,
+      "eval_runtime": 9.4961,
+      "eval_samples_per_second": 21.061,
+      "eval_steps_per_second": 2.633,
+      "step": 960
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 0.00013686967224971273,
+      "loss": 1.4942,
+      "step": 965
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 0.00013588187023566163,
+      "loss": 1.782,
+      "step": 970
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 0.00013424610703122958,
+      "loss": 1.8084,
+      "step": 975
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 0.00013197813593027432,
+      "loss": 1.7091,
+      "step": 980
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 0.00012909979873429724,
+      "loss": 1.8472,
+      "step": 985
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 0.00012563881540395458,
+      "loss": 1.7417,
+      "step": 990
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 0.00012162851710068373,
+      "loss": 1.6717,
+      "step": 995
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 0.00011710752518939715,
+      "loss": 1.6891,
+      "step": 1000
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 0.00011211937929362609,
+      "loss": 1.8756,
+      "step": 1005
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 0.00010671211798514474,
+      "loss": 1.6079,
+      "step": 1010
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 0.00010093781614626349,
+      "loss": 1.7571,
+      "step": 1015
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 9.485208346024524e-05,
+      "loss": 1.3666,
+      "step": 1020
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 8.851352885965622e-05,
+      "loss": 1.7333,
+      "step": 1025
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 8.198319609030653e-05,
+      "loss": 1.4846,
+      "step": 1030
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 7.532397582660802e-05,
+      "loss": 1.4929,
+      "step": 1035
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 6.859999999999993e-05,
+      "loss": 1.7269,
+      "step": 1040
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 6.18760241733921e-05,
+      "loss": 1.5586,
+      "step": 1045
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 5.5216803909693576e-05,
+      "loss": 1.7124,
+      "step": 1050
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 4.868647114034389e-05,
+      "loss": 1.7847,
+      "step": 1055
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 4.2347916539754865e-05,
+      "loss": 1.5791,
+      "step": 1060
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 3.62621838537366e-05,
+      "loss": 1.71,
+      "step": 1065
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 3.0487882014855342e-05,
+      "loss": 1.6788,
+      "step": 1070
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.5080620706373995e-05,
+      "loss": 1.6755,
+      "step": 1075
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 2.009247481060292e-05,
+      "loss": 1.4464,
+      "step": 1080
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 1.5571482899316333e-05,
+      "loss": 1.7238,
+      "step": 1085
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 1.1561184596045489e-05,
+      "loss": 1.4748,
+      "step": 1090
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 8.100201265702821e-06,
+      "loss": 1.488,
+      "step": 1095
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 5.221864069725715e-06,
+      "loss": 1.545,
+      "step": 1100
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.9538929687704672e-06,
+      "loss": 1.5979,
+      "step": 1105
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 1.3181297643383925e-06,
+      "loss": 1.7046,
+      "step": 1110
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 3.303277502872983e-07,
+      "loss": 1.5848,
+      "step": 1115
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0,
+      "loss": 1.6901,
+      "step": 1120
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.4377235174179077,
+      "eval_runtime": 9.5035,
+      "eval_samples_per_second": 21.045,
+      "eval_steps_per_second": 2.631,
+      "step": 1120
     }
   ],
-  "max_steps": 790,
-  "num_train_epochs": 5,
-  "total_flos": 825029591040000.0,
+  "max_steps": 1120,
+  "num_train_epochs": 7,
+  "total_flos": 1168759259136000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cfb6e247e098310914adb6ffa99cbac18054688ed991eef0c2acb3531d8319c
-size 2863
+oid sha256:1ca623b70d31f90284e0c9f8a31f7cb178f04870d48cd416af0fe84dfe57c1e5
+size 3055