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/8f3967015c2cde3fec618068eb3f07ae.1000x1000x1.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/b12dc90e6f405684ef6b74c9de92fdcd.853x853x1.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/platina")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3djk6dl5/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/ugrse2us/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 Платина (Platina)'s lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/36yxpve3) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/174snup8) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/36yxpve3/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/174snup8/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/platina",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -35,7 +35,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.10.0",
+  "transformers_version": "4.11.1",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 1.7263113260269165, "eval_runtime": 6.8067, "eval_samples_per_second": 21.009, "eval_steps_per_second": 2.644, "epoch": 1.0}
+{"eval_loss": 1.5137468576431274, "eval_runtime": 5.9741, "eval_samples_per_second": 21.928, "eval_steps_per_second": 2.846, "epoch": 4.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60483dee43c8e75c58020f148a3baba445919cfda40b404f1b56f03b01867fb5
+oid sha256:674bcbe4c3f0b74c808b2acddd4c29980ae0e900a2d265df4a57bdda5b5995cb
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5231c799b2a746345dd765f04875b87a9e5b984055d1489b083801b9972ebd36
-size 995603825
+oid sha256:6c84c8d034b21501bcaa1bb5445c0df2a821aa1846fc23438d5d5b489ae916db
+size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:769d11bf0271156105c2d290ba6fc0d4571c986b852aa0ef0a29f47c140fba01
+oid sha256:dfad19679cf041d3a6d0d3c84576e0a908395320dd1342c1eff0220a5cbf7829
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87ebaa6c977ec62835f08aadd3b506bee1df391c1e6d9f1b6aaee1266273a90e
-size 14503
+oid sha256:793f8e26475d3635b181b3b379f08c9cf76d34985095cf8466590d2e2b8b1156
+size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2442c0e486037041bbb3784e336cf93be1cf199df3328eb674c20fce504cbae
+oid sha256:ca7960f9c01bdbbfc38037b14cc37d37af713a0aaaedd217788c1312c4a9e32c
 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/platina", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,126 +1,462 @@
 {
-  "best_metric": 1.7263113260269165,
-  "best_model_checkpoint": "output/platina/checkpoint-86",
-  "epoch": 1.0,
-  "global_step": 86,
+  "best_metric": 1.5137468576431274,
+  "best_model_checkpoint": "output/platina/checkpoint-348",
+  "epoch": 4.0,
+  "global_step": 348,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.06,
-      "learning_rate": 0.00013605888543050137,
-      "loss": 2.653,
+      "learning_rate": 1.1151037062646087e-06,
+      "loss": 1.8191,
       "step": 5
     },
     {
-      "epoch": 0.12,
-      "learning_rate": 0.0001326735050590528,
-      "loss": 2.339,
+      "epoch": 0.11,
+      "learning_rate": 4.424162455503776e-06,
+      "loss": 1.778,
       "step": 10
     },
     {
       "epoch": 0.17,
-      "learning_rate": 0.00012715648590796197,
-      "loss": 2.2005,
+      "learning_rate": 9.819597714903368e-06,
+      "loss": 1.7677,
       "step": 15
     },
     {
       "epoch": 0.23,
-      "learning_rate": 0.00011969137173622343,
-      "loss": 2.1397,
+      "learning_rate": 1.7126002200070792e-05,
+      "loss": 1.8796,
       "step": 20
     },
     {
       "epoch": 0.29,
-      "learning_rate": 0.00011052651678711246,
-      "loss": 2.05,
+      "learning_rate": 2.610584242060141e-05,
+      "loss": 1.8322,
       "step": 25
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 9.99668233786982e-05,
-      "loss": 1.9257,
+      "epoch": 0.34,
+      "learning_rate": 3.646718096799441e-05,
+      "loss": 1.7206,
       "step": 30
     },
     {
-      "epoch": 0.41,
-      "learning_rate": 8.83635982164553e-05,
-      "loss": 1.9766,
+      "epoch": 0.4,
+      "learning_rate": 4.787316749192729e-05,
+      "loss": 1.7349,
       "step": 35
     },
     {
-      "epoch": 0.47,
-      "learning_rate": 7.61028648947222e-05,
-      "loss": 1.9569,
+      "epoch": 0.46,
+      "learning_rate": 5.995298981028814e-05,
+      "loss": 1.7768,
       "step": 40
     },
     {
       "epoch": 0.52,
-      "learning_rate": 6.359252141426174e-05,
-      "loss": 1.9599,
+      "learning_rate": 7.231392912895959e-05,
+      "loss": 1.7442,
       "step": 45
     },
     {
-      "epoch": 0.58,
-      "learning_rate": 5.1248769967930697e-05,
-      "loss": 1.8944,
+      "epoch": 0.57,
+      "learning_rate": 8.455412745239765e-05,
+      "loss": 1.7635,
       "step": 50
     },
     {
-      "epoch": 0.64,
-      "learning_rate": 3.9482270457119095e-05,
-      "loss": 1.8734,
+      "epoch": 0.63,
+      "learning_rate": 9.627565211250085e-05,
+      "loss": 1.7607,
       "step": 55
     },
     {
-      "epoch": 0.7,
-      "learning_rate": 2.8684478392724447e-05,
-      "loss": 1.8827,
+      "epoch": 0.69,
+      "learning_rate": 0.00010709743268385943,
+      "loss": 1.8494,
       "step": 60
     },
     {
-      "epoch": 0.76,
-      "learning_rate": 1.9214621700201495e-05,
-      "loss": 1.8779,
+      "epoch": 0.75,
+      "learning_rate": 0.0001166676497022313,
+      "loss": 1.7513,
       "step": 65
     },
     {
-      "epoch": 0.81,
-      "learning_rate": 1.1387749692754382e-05,
-      "loss": 1.776,
+      "epoch": 0.8,
+      "learning_rate": 0.00012467517242513996,
+      "loss": 1.6916,
       "step": 70
     },
     {
-      "epoch": 0.87,
-      "learning_rate": 5.464251807155671e-06,
-      "loss": 1.8927,
+      "epoch": 0.86,
+      "learning_rate": 0.00013085967378942764,
+      "loss": 1.8089,
       "step": 75
     },
     {
-      "epoch": 0.93,
-      "learning_rate": 1.6411947996634035e-06,
-      "loss": 1.8982,
+      "epoch": 0.92,
+      "learning_rate": 0.00013502009372533583,
+      "loss": 1.7891,
       "step": 80
     },
     {
-      "epoch": 0.99,
-      "learning_rate": 4.576660180157781e-08,
-      "loss": 1.8376,
+      "epoch": 0.98,
+      "learning_rate": 0.00013702117568213136,
+      "loss": 1.8074,
       "step": 85
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.7263113260269165,
-      "eval_runtime": 6.7602,
-      "eval_samples_per_second": 21.153,
-      "eval_steps_per_second": 2.663,
-      "step": 86
+      "eval_loss": 1.717990517616272,
+      "eval_runtime": 5.9537,
+      "eval_samples_per_second": 22.003,
+      "eval_steps_per_second": 2.855,
+      "step": 87
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00013679786386078908,
+      "loss": 1.6816,
+      "step": 90
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00013435741819832448,
+      "loss": 1.6466,
+      "step": 95
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00012977917834497524,
+      "loss": 1.7089,
+      "step": 100
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00012321198430740717,
+      "loss": 1.6398,
+      "step": 105
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00011486933761363929,
+      "loss": 1.5449,
+      "step": 110
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00010502246031172626,
+      "loss": 1.6906,
+      "step": 115
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.399147745631814e-05,
+      "loss": 1.5677,
+      "step": 120
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 8.213500974321104e-05,
+      "loss": 1.6043,
+      "step": 125
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 6.983851463858842e-05,
+      "loss": 1.658,
+      "step": 130
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 5.750175503648031e-05,
+      "loss": 1.6382,
+      "step": 135
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.5525802842298056e-05,
+      "loss": 1.5709,
+      "step": 140
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.4300000000000054e-05,
+      "loss": 1.6331,
+      "step": 145
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.418930086396666e-05,
+      "loss": 1.5709,
+      "step": 150
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.552240741902345e-05,
+      "loss": 1.5756,
+      "step": 155
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 8.581083076314943e-06,
+      "loss": 1.6289,
+      "step": 160
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.5909924568597513e-06,
+      "loss": 1.5374,
+      "step": 165
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 7.143649642832784e-07,
+      "loss": 1.5678,
+      "step": 170
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.5926672220230103,
+      "eval_runtime": 5.7532,
+      "eval_samples_per_second": 22.77,
+      "eval_steps_per_second": 2.955,
+      "step": 174
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4720656268161284e-08,
+      "loss": 1.6048,
+      "step": 175
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 1.603829878288063e-06,
+      "loss": 1.4474,
+      "step": 180
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 5.341005502881008e-06,
+      "loss": 1.5889,
+      "step": 185
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 1.1134750783984455e-05,
+      "loss": 1.4795,
+      "step": 190
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.87967092540732e-05,
+      "loss": 1.5209,
+      "step": 195
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.8077788251711776e-05,
+      "loss": 1.6414,
+      "step": 200
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 3.8676257001728066e-05,
+      "loss": 1.5243,
+      "step": 205
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 5.0247555976895465e-05,
+      "loss": 1.5045,
+      "step": 210
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 6.241549863571236e-05,
+      "loss": 1.4689,
+      "step": 215
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 7.47845013642876e-05,
+      "loss": 1.5789,
+      "step": 220
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 8.695244402310448e-05,
+      "loss": 1.5342,
+      "step": 225
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 9.852374299827188e-05,
+      "loss": 1.4686,
+      "step": 230
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.00010912221174828818,
+      "loss": 1.5636,
+      "step": 235
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.00011840329074592676,
+      "loss": 1.5472,
+      "step": 240
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.00012606524921601552,
+      "loss": 1.5416,
+      "step": 245
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00013185899449711896,
+      "loss": 1.5624,
+      "step": 250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00013559617012171192,
+      "loss": 1.5904,
+      "step": 255
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.00013715527934373185,
+      "loss": 1.5604,
+      "step": 260
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.5875556468963623,
+      "eval_runtime": 5.7815,
+      "eval_samples_per_second": 22.659,
+      "eval_steps_per_second": 2.94,
+      "step": 261
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 0.0001364856350357167,
+      "loss": 1.5918,
+      "step": 265
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 0.0001336090075431402,
+      "loss": 1.4197,
+      "step": 270
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 0.00012861891692368514,
+      "loss": 1.5324,
+      "step": 275
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 0.00012167759258097666,
+      "loss": 1.4255,
+      "step": 280
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 0.00011301069913603348,
+      "loss": 1.5272,
+      "step": 285
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 0.00010290000000000009,
+      "loss": 1.52,
+      "step": 290
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 9.167419715770211e-05,
+      "loss": 1.4541,
+      "step": 295
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 7.969824496351975e-05,
+      "loss": 1.4829,
+      "step": 300
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 6.736148536141163e-05,
+      "loss": 1.4054,
+      "step": 305
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 5.506499025678901e-05,
+      "loss": 1.4169,
+      "step": 310
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 4.320852254368191e-05,
+      "loss": 1.4294,
+      "step": 315
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 3.217753968827379e-05,
+      "loss": 1.4879,
+      "step": 320
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 2.2330662386360752e-05,
+      "loss": 1.4319,
+      "step": 325
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 1.3988015692592839e-05,
+      "loss": 1.4294,
+      "step": 330
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 7.420821655024771e-06,
+      "loss": 1.4842,
+      "step": 335
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 2.8425818016755186e-06,
+      "loss": 1.3857,
+      "step": 340
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 4.02136139210924e-07,
+      "loss": 1.3751,
+      "step": 345
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.5137468576431274,
+      "eval_runtime": 5.7841,
+      "eval_samples_per_second": 22.648,
+      "eval_steps_per_second": 2.939,
+      "step": 348
     }
   ],
-  "max_steps": 86,
-  "num_train_epochs": 1,
-  "total_flos": 88969936896000.0,
+  "max_steps": 348,
+  "num_train_epochs": 4,
+  "total_flos": 362150756352000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:184b4ca1c322207d846f9c8c7d8c7a0e10241a7f02b04480ce0d867908a4103a
-size 2671
+oid sha256:d29bc07a887bc66db2f72baab7071a8ed663e30bb50cd30619db8b1114d32a09
+size 2863