huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/morgenshtern")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2mqz06ef/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/26pogqcg/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 MORGENSHTERN's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1u21f1zw) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2f54nbz5) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1u21f1zw/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2f54nbz5/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -35,7 +35,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.10.1",
+  "transformers_version": "4.10.2",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 1.029100775718689, "eval_runtime": 7.2168, "eval_samples_per_second": 20.508, "eval_steps_per_second": 2.633, "epoch": 21.0}
+{"eval_loss": 1.004758358001709, "eval_runtime": 6.3083, "eval_samples_per_second": 20.766, "eval_steps_per_second": 2.695, "epoch": 14.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22fb23883fc29f2e2299e376a3685d3899aa27180ebd009e6a276dd79c0f549a
+oid sha256:73f4525a915f9e923ec8c1948b7703789a43fc65ae0c0ce7a70b29cc32f99056
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6cae6db0edbbd42bbc453b0149987a6ee3a6ba790c60e2925ab382f8c1f160ba
+oid sha256:e3189c4875db3d6c2430dd21cc7aaf58494cb3ea16a1644a9eba14559109a180
 size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0d244ecace2e78de24953f6135d7cc932ff6489b68a4c749677efd661e57029
+oid sha256:77256682836976672d4132cd08e64ed8578aca99003f5b1ad793033c0c217d56
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76d31626c3bac6215c579c04c793834783bc9d247d3009c55526f8f7392b8d8e
-size 14567
+oid sha256:c58bc32277fe141391df912d7cedc705240c20ad9546d00ca7addc7fbb5b47a9
+size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d04e24471eba096654369d2172acef6ba07c54938134ee47e6aca1b621eca74
+oid sha256:7618b7dada376a378c38d9aa39cd28c5eded00be56339f85d3ea9724b270283c
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 1.029100775718689,
-  "best_model_checkpoint": "output/morgenshtern/checkpoint-1274",
+  "best_metric": 1.004758358001709,
+  "best_model_checkpoint": "output/morgenshtern/checkpoint-1300",
   "epoch": 13.0,
-  "global_step": 1274,
+  "global_step": 1300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1674,11 +1674,55 @@
       "eval_samples_per_second": 21.242,
       "eval_steps_per_second": 2.727,
       "step": 1274
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 2.0092474810602958e-05,
+      "loss": 1.2194,
+      "step": 1275
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.3101434185878674e-05,
+      "loss": 1.0397,
+      "step": 1280
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 7.476952440677985e-06,
+      "loss": 1.0537,
+      "step": 1285
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 3.35752298215246e-06,
+      "loss": 1.2597,
+      "step": 1290
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 8.445798351736176e-07,
+      "loss": 1.0949,
+      "step": 1295
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 0.0,
+      "loss": 1.1523,
+      "step": 1300
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 1.004758358001709,
+      "eval_runtime": 6.2971,
+      "eval_samples_per_second": 20.803,
+      "eval_steps_per_second": 2.7,
+      "step": 1300
     }
   ],
-  "max_steps": 2058,
-  "num_train_epochs": 21,
-  "total_flos": 1320047345664000.0,
+  "max_steps": 1400,
+  "num_train_epochs": 14,
+  "total_flos": 1346568486912000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:188b9cc77247d315e833007ec5b6ee4bead6e01dfa5781ac98b51d69684404d5
+oid sha256:5e678611f04c51489e9a36568716af96aebefa9660b4f8c5449f0a99f9629d37
 size 2671