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/261aeb8c50a24a787fb1b8e5ba4fa356.1000x1000x1.png&#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/29cedf8dd30a7458f4fca47d1c0f0eab.1000x1000x1.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/rammstein")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2c3w0gbx/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/qt3qa1x1/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 Rammstein's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/19sol366) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2yyigjzv) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/19sol366/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2yyigjzv/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "rammstein",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -36,7 +36,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.12.5",
+  "transformers_version": "4.19.2",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.5774738788604736, "eval_runtime": 2.4486, "eval_samples_per_second": 21.645, "eval_steps_per_second": 2.859, "epoch": 1.0}
+{"eval_loss": 2.3011960983276367, "eval_runtime": 0.7541, "eval_samples_per_second": 79.564, "eval_steps_per_second": 10.609, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c50a76fcc42d6af60cc2e50d89dda9661511c20eceea8252ad71d880069c2fb
+oid sha256:03c08e945202ddf1ebd8348598751d6c196aa56293a3859e6127461b5c6be2ad
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b495868c47b36fe332edc517898b146b83d2efe864bf6a95d52df5c5fff423f
+oid sha256:cebc41cae62202846215c2c1720dc6e09f5c67ecb6bfadaf587db41eb94d3086
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93b59c1dcfc021f0eb24592c8c12572ef53d87235821786c53cc766d55115f3a
-size 510403817
+oid sha256:e338ab740793bb9a9e4aac9c5c91021b616ca47dba2e7f98c96b90f92da75a93
+size 510396521

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff5122ba1d15b1ef4fb1719947ed61b9d648befcfcad85f66a3ef938ed8ad41a
+oid sha256:5fa5222bb5aaf4570a2485f42de06ccf0e38362fb68506b4fa8d9280490a1a50
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b336f900ee7256719f9a9e8b26a06137b0cfb005089a677cbd7e3a7ec3f58f39
+oid sha256:12ce1132cbb6aa5c18263e33796f9990b4148f45c42e80bbd2185f1d18ecd3d7
 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/rammstein", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,66 +1,146 @@
 {
-  "best_metric": 2.5774738788604736,
-  "best_model_checkpoint": "output/rammstein/checkpoint-36",
-  "epoch": 1.0,
-  "global_step": 36,
+  "best_metric": 2.3011960983276367,
+  "best_model_checkpoint": "output/rammstein/checkpoint-96",
+  "epoch": 2.0,
+  "global_step": 96,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.14,
-      "learning_rate": 0.0001307727141907142,
-      "loss": 3.542,
+      "epoch": 0.1,
+      "learning_rate": 3.64059311663575e-06,
+      "loss": 2.9482,
       "step": 5
     },
     {
-      "epoch": 0.28,
-      "learning_rate": 0.00011269523002449659,
-      "loss": 3.2347,
+      "epoch": 0.21,
+      "learning_rate": 1.4175960856021253e-05,
+      "loss": 2.7279,
       "step": 10
     },
     {
-      "epoch": 0.42,
-      "learning_rate": 8.635498649403293e-05,
-      "loss": 3.1198,
+      "epoch": 0.31,
+      "learning_rate": 3.048788201485529e-05,
+      "loss": 3.0408,
       "step": 15
     },
     {
-      "epoch": 0.56,
-      "learning_rate": 5.668773501204858e-05,
-      "loss": 2.8512,
+      "epoch": 0.42,
+      "learning_rate": 5.084501350596709e-05,
+      "loss": 2.8174,
       "step": 20
     },
     {
-      "epoch": 0.69,
-      "learning_rate": 2.9252656466318256e-05,
-      "loss": 2.8823,
+      "epoch": 0.52,
+      "learning_rate": 7.308665466518777e-05,
+      "loss": 2.9266,
       "step": 25
     },
     {
-      "epoch": 0.83,
-      "learning_rate": 9.190657300387505e-06,
-      "loss": 3.015,
+      "epoch": 0.62,
+      "learning_rate": 9.485208346024518e-05,
+      "loss": 2.7754,
       "step": 30
     },
     {
-      "epoch": 0.97,
-      "learning_rate": 2.610437109062556e-07,
-      "loss": 3.1043,
+      "epoch": 0.73,
+      "learning_rate": 0.00011383112291586474,
+      "loss": 2.4825,
       "step": 35
     },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00012800934269961248,
+      "loss": 2.6286,
+      "step": 40
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0001358818702356616,
+      "loss": 2.681,
+      "step": 45
+    },
     {
       "epoch": 1.0,
-      "eval_loss": 2.5774738788604736,
-      "eval_runtime": 2.4107,
-      "eval_samples_per_second": 21.985,
-      "eval_steps_per_second": 2.904,
-      "step": 36
+      "eval_loss": 2.4657435417175293,
+      "eval_runtime": 0.7344,
+      "eval_samples_per_second": 81.697,
+      "eval_steps_per_second": 10.893,
+      "step": 48
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0001366131174902434,
+      "loss": 2.5211,
+      "step": 50
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0001301254700691424,
+      "loss": 2.3708,
+      "step": 55
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00011710752518939717,
+      "loss": 2.4006,
+      "step": 60
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.894100414902355e-05,
+      "loss": 2.3993,
+      "step": 65
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 7.755409678629555e-05,
+      "loss": 2.4798,
+      "step": 70
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.5216803909693664e-05,
+      "loss": 2.4973,
+      "step": 75
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.4300000000000054e-05,
+      "loss": 2.2237,
+      "step": 80
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7023789206942107e-05,
+      "loss": 2.2823,
+      "step": 85
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 5.221864069725753e-06,
+      "loss": 2.2183,
+      "step": 90
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.4687786583180135e-07,
+      "loss": 2.2238,
+      "step": 95
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.3011960983276367,
+      "eval_runtime": 0.7368,
+      "eval_samples_per_second": 81.434,
+      "eval_steps_per_second": 10.858,
+      "step": 96
     }
   ],
-  "max_steps": 36,
-  "num_train_epochs": 1,
-  "total_flos": 37234114560000.0,
+  "max_steps": 96,
+  "num_train_epochs": 2,
+  "total_flos": 99552264192000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:def40346d44df4f059692ae0bbbf8eb1abeaa9f743f81a40c75e899ae4b9d3b0
-size 2863
+oid sha256:21bd930efff238a97f2bba32a2ac5d18401b5fa862717f400510445f4a9addf4
+size 3247