huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/linkin-park")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1s6xt77z/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/2b15vg01/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 Linkin Park's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1i4je18p) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/fto51dp4) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1i4je18p/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/fto51dp4/artifacts) is logged and versioned.
 
 ## How to use
 

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.248004913330078, "eval_runtime": 4.6644, "eval_samples_per_second": 20.796, "eval_steps_per_second": 2.787, "epoch": 2.0}
+{"eval_loss": 2.0318267345428467, "eval_runtime": 4.5697, "eval_samples_per_second": 21.008, "eval_steps_per_second": 2.626, "epoch": 3.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:225d151b89706812d44a3b7db2ec9b070074c0f7ba7e9d0ffe24145cc0990adb
+oid sha256:960e645cfdb5661c8a5575898191f3f17c064dc742a6f5d1c1e623c5a21c709e
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35bd95469cbf0ab080be66768e5242378073c1190b28bd15f2700cf295d526f6
+oid sha256:f8e18e8cd6f8f5633cbbd1851d80023da278856308536afdaf38cbfb154a8732
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a3eb8db73abd6dfffd1f85d4cbb59cf5647bfbe01d3c02e21a5a72a96906a3a
+oid sha256:a568564c9d7a3f53afa0d2c83e7145788515ff74c6c201a2bf1f4cce94aecb72
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db7af3f0fec071276d1a4d8600845adbb6ac8e80806a422d4311d79890098c71
+oid sha256:623c33d8f94c376806d8d4f4fd33e1db6afbc9689cfdf6d4fc7c47fa87f23bdd
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03b8c59abd7807eb31fbb6b182b880936157d534c8d17e22038fc44702f9354e
+oid sha256:81c6de287d0aeee2c682cd2a01e5e01657f181e757ee594a86aa0945f540d0b1
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.248004913330078,
-  "best_model_checkpoint": "output/linkin-park/checkpoint-140",
-  "epoch": 2.0,
-  "global_step": 140,
+  "best_metric": 2.0318267345428467,
+  "best_model_checkpoint": "output/linkin-park/checkpoint-210",
+  "epoch": 3.0,
+  "global_step": 210,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -198,11 +198,103 @@
       "eval_samples_per_second": 21.058,
       "eval_steps_per_second": 2.822,
       "step": 140
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.0001354800547756731,
+      "loss": 2.1018,
+      "step": 145
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00013040646433810595,
+      "loss": 1.8963,
+      "step": 150
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00012223363969730686,
+      "loss": 2.1895,
+      "step": 155
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.00011137140040750914,
+      "loss": 2.0564,
+      "step": 160
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 9.836442450346452e-05,
+      "loss": 2.1397,
+      "step": 165
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 8.386493606940326e-05,
+      "loss": 2.3483,
+      "step": 170
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 6.860000000000001e-05,
+      "loss": 1.9349,
+      "step": 175
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 5.333506393059685e-05,
+      "loss": 1.8439,
+      "step": 180
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 3.8835575496535535e-05,
+      "loss": 1.9104,
+      "step": 185
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 2.58285995924909e-05,
+      "loss": 1.7206,
+      "step": 190
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 1.496636030269317e-05,
+      "loss": 2.2577,
+      "step": 195
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 6.793535661894062e-06,
+      "loss": 1.8714,
+      "step": 200
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.7199452243269073e-06,
+      "loss": 2.2299,
+      "step": 205
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 1.8604,
+      "step": 210
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.0318267345428467,
+      "eval_runtime": 4.5539,
+      "eval_samples_per_second": 21.081,
+      "eval_steps_per_second": 2.635,
+      "step": 210
     }
   ],
-  "max_steps": 140,
-  "num_train_epochs": 2,
-  "total_flos": 144625139712000.0,
+  "max_steps": 210,
+  "num_train_epochs": 3,
+  "total_flos": 217264324608000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c394c879763a7ac8c43760c1791f5a9aee9dc1f21c84e74abc5a6b80af9588a
+oid sha256:aada2de1d92f467f8f5669200ff659c45b17df03bf26159792e731f4a3fea9f2
 size 2671