huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/aaron-watson")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/204aifr7/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/14ha1tnc/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 Aaron Watson's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/20vl7rt5) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/34e4zb2v) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/20vl7rt5/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/34e4zb2v/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -35,7 +35,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.10.0",
+  "transformers_version": "4.10.1",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.9697048664093018, "eval_runtime": 1.8075, "eval_samples_per_second": 20.47, "eval_steps_per_second": 2.766, "epoch": 5.0}
+{"eval_loss": 2.5737111568450928, "eval_runtime": 1.6076, "eval_samples_per_second": 20.527, "eval_steps_per_second": 3.11, "epoch": 13.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75c4ed6e042b687530615246e336ead94b468f8bffd1f840eeb3535c66e14912
+oid sha256:230fc6b9831af6a5d461ab334806ff82700da38ca04f9db2a0322a2e5602d013
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af00edd8315e8240fc11cc4bbe4fd8812f8e9944d63e1e804ed4135ef7995423
+oid sha256:c38111d201ac21b411943ec105908899853a9b846ba960bdcb57446aef22532f
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ae3248a48d48f36b9ecb728e6afed45c93ad66bd3f5adf5a346cb1e380479b5
+oid sha256:f1322e8004a848bba72952c47a4a02987aee17b648c9f98341279d603dd9a5bb
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6e12081284488e0302f0f15cc7b9e867b93ab823a7f5b31fa14b05b508db300
-size 14503
+oid sha256:8f05d20325eb8102d8b4c3e4d9d799ee273f1bf0946e095b1a0f1f287bd5eb85
+size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d625cddcadf735c6fc48a4f3f1d506c16673a1b6749611364d76e3a20b91fce
+oid sha256:7358664183e7649e0e2f607694e0c72ac495305b3a893efe8d2bc60b5501be46
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.9697048664093018,
-  "best_model_checkpoint": "output/aaron-watson/checkpoint-75",
-  "epoch": 3.0,
-  "global_step": 75,
+  "best_metric": 2.5737111568450928,
+  "best_model_checkpoint": "output/aaron-watson/checkpoint-100",
+  "epoch": 4.0,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -120,11 +120,49 @@
       "eval_samples_per_second": 21.208,
       "eval_steps_per_second": 2.866,
       "step": 75
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 1.310143418587859e-05,
+      "loss": 2.8552,
+      "step": 80
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 4.740143418587858e-05,
+      "loss": 2.8477,
+      "step": 85
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 8.979856581412137e-05,
+      "loss": 2.825,
+      "step": 90
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 2.7383,
+      "step": 95
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0001372,
+      "loss": 2.839,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.5737111568450928,
+      "eval_runtime": 1.5782,
+      "eval_samples_per_second": 20.911,
+      "eval_steps_per_second": 3.168,
+      "step": 100
     }
   ],
-  "max_steps": 125,
-  "num_train_epochs": 5,
-  "total_flos": 76689211392000.0,
+  "max_steps": 325,
+  "num_train_epochs": 13,
+  "total_flos": 102557122560000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66237f886a59d41c0e749b69ee85bc3f1c09d6f131ed3be21ee7e5ab55356657
+oid sha256:46292e67196f36dc6e1d54180d712db3c13701f4c597d47be75b9ef2cadcf193
 size 2671