huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/drake")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1aksz894/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/2e42ok17/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 Drake's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3rx3sf99) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2xe72oq3) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3rx3sf99/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2xe72oq3/artifacts) is logged and versioned.
 
 ## How to use
 

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.985200881958008, "eval_runtime": 3.2437, "eval_samples_per_second": 143.353, "eval_steps_per_second": 18.189, "epoch": 12.0}
+{"eval_loss": 2.6586830615997314, "eval_runtime": 12.618, "eval_samples_per_second": 43.906, "eval_steps_per_second": 5.548, "epoch": 6.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57a832a766a6062bfcfcaea51111be32f951d3ccc6a07c44a1c70945134909a5
+oid sha256:bb6ff1968e33bafe123ccca386bcb5b25cdd34682f12236ab1c25d04630b79f6
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc861593254336ef6f7349effb09e8d7559406594cf64d7e16e4803c44118418
+oid sha256:de10120cd339dc8a2411ec8e4b7e1d4af7983ab863a59e4fb699101f393fde92
 size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0b2b81b7c939397be91b9a719857e4d86a375656547375710d1aeb2afa5ee16
+oid sha256:4cce53ebbc5d19a5a3d4fe01b3d7363a3a0c4f9cb9bdafea201bad39efd63457
 size 510396521

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f37b1fe21c3237da257fc4dad062a40dd556ee8402c38632ba55aeb299ee646a
+oid sha256:326cc7de00c3f20c4fcf0338fdce1a6ac81da32e3abb8094e79488b6be2ca9fb
 size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcfeaed1730d77e101fd5ed4b37bb1cb254cb585fa58085f4a7ca52a63e8aab8
+oid sha256:3ae78cbe0368ef751528f9df25bdcc580eaf09b6340910f9e0e6aae733d99451
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.985200881958008,
-  "best_model_checkpoint": "output/drake/checkpoint-1640",
+  "best_metric": 2.6586830615997314,
+  "best_model_checkpoint": "output/drake/checkpoint-1755",
   "epoch": 5.0,
-  "global_step": 1640,
+  "global_step": 1755,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2006,11 +2006,157 @@
       "eval_samples_per_second": 142.792,
       "eval_steps_per_second": 18.118,
       "step": 1640
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.064756856761554e-05,
+      "loss": 2.7681,
+      "step": 1645
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 2.8129056814413798e-05,
+      "loss": 2.7734,
+      "step": 1650
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 2.5691584488661662e-05,
+      "loss": 2.8042,
+      "step": 1655
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 2.3340032409686368e-05,
+      "loss": 2.8781,
+      "step": 1660
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 2.1079109349241507e-05,
+      "loss": 2.8162,
+      "step": 1665
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 1.891334260261341e-05,
+      "loss": 2.523,
+      "step": 1670
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 1.6847068923119758e-05,
+      "loss": 2.8495,
+      "step": 1675
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 1.4884425838151134e-05,
+      "loss": 2.791,
+      "step": 1680
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 1.3029343364143164e-05,
+      "loss": 2.6161,
+      "step": 1685
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 1.1285536137072603e-05,
+      "loss": 2.8099,
+      "step": 1690
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 9.65649597423201e-06,
+      "loss": 2.9416,
+      "step": 1695
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 8.145484882179486e-06,
+      "loss": 2.881,
+      "step": 1700
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 6.755528524864052e-06,
+      "loss": 2.4485,
+      "step": 1705
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 5.4894101650047195e-06,
+      "loss": 2.7043,
+      "step": 1710
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 4.349665090857373e-06,
+      "loss": 2.9465,
+      "step": 1715
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 3.3385755395280496e-06,
+      "loss": 2.8094,
+      "step": 1720
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 2.4581661269975954e-06,
+      "loss": 2.8454,
+      "step": 1725
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 1.7101997940107123e-06,
+      "loss": 2.717,
+      "step": 1730
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 1.0961742759449984e-06,
+      "loss": 2.6812,
+      "step": 1735
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 6.173191037306866e-07,
+      "loss": 2.7196,
+      "step": 1740
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 2.7459314182545106e-07,
+      "loss": 2.8512,
+      "step": 1745
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 6.868266817434572e-08,
+      "loss": 2.8381,
+      "step": 1750
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 2.7245,
+      "step": 1755
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6586830615997314,
+      "eval_runtime": 12.7857,
+      "eval_samples_per_second": 43.33,
+      "eval_steps_per_second": 5.475,
+      "step": 1755
     }
   ],
-  "max_steps": 3936,
-  "num_train_epochs": 12,
-  "total_flos": 1711070871552000.0,
+  "max_steps": 2106,
+  "num_train_epochs": 6,
+  "total_flos": 1831003914240000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05834366ef55540d4e60793dd81a674c9d97225933a6dddaf453b9a50205534e
+oid sha256:7b14d0e1bec474f2c0b3b9e74e87729a4504a63316afeed249b70817f8d53a13
 size 3311