huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/nirvana")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/plmwg217/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/1bj9eav1/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 Nirvana's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2aaxwzd4) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3vzztlsq) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2aaxwzd4/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3vzztlsq/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -37,7 +37,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.16.1",
+  "transformers_version": "4.16.2",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 1.7358663082122803, "eval_runtime": 2.0349, "eval_samples_per_second": 20.64, "eval_steps_per_second": 2.949, "epoch": 3.0}
+{"eval_loss": 1.652644395828247, "eval_runtime": 0.6861, "eval_samples_per_second": 72.881, "eval_steps_per_second": 10.203, "epoch": 4.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38b8911734a29e325912427efb96ad710eeebdec940f173c4c1cdf4bd6295893
+oid sha256:bb7ffe29d6c9df8eb6f354e7594b14e64327efbd1282dcc6bfb98157a785df95
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ea53252af8c636e7d87ccfe4ffa643cb49756436eab6186e4214ea668b415a6
+oid sha256:ce5b25f796f3ce284994f482729b66cbae756dbc6c6713fa0e06c01020286111
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75c2f6598fea7212344467f26fdd0991f32656d8a34cf5910e5c27adb467fa3d
+oid sha256:eba841c32d0a8e1b4d077ed2c72a586a4769ea7b8bcd773668d5638993a4391a
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:379e3990b5cff526d3b72fba15374693b30ed8e1821d3514ef232ad47d9ec9f7
+oid sha256:ef49b6bb7686a0d942a75007f3b3fa50bd852d104828a62e770eaeb6558f5b5b
 size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5704cdbfa3a8ad0fca4ee30e2feb20b3d392393975fb15db2691fbf243dcbce0
+oid sha256:90b218d2fa339ad9451b337b3dfa7239e2fc5fb06e4d1e48c52d8dfd833c0e23
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 1.7358663082122803,
-  "best_model_checkpoint": "output/nirvana/checkpoint-111",
-  "epoch": 3.0,
-  "global_step": 111,
+  "best_metric": 1.652644395828247,
+  "best_model_checkpoint": "output/nirvana/checkpoint-144",
+  "epoch": 4.0,
+  "global_step": 144,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -178,11 +178,55 @@
       "eval_samples_per_second": 22.729,
       "eval_steps_per_second": 3.247,
       "step": 111
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.2406169761775193e-05,
+      "loss": 1.9469,
+      "step": 115
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.4300000000000014e-05,
+      "loss": 1.7381,
+      "step": 120
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 6.262111604751063e-05,
+      "loss": 1.8423,
+      "step": 125
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 9.206258183214083e-05,
+      "loss": 1.6384,
+      "step": 130
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 0.0001171075251893971,
+      "loss": 2.0094,
+      "step": 135
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 0.00013306291378591332,
+      "loss": 1.8349,
+      "step": 140
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.652644395828247,
+      "eval_runtime": 0.6674,
+      "eval_samples_per_second": 74.92,
+      "eval_steps_per_second": 10.489,
+      "step": 144
     }
   ],
-  "max_steps": 111,
-  "num_train_epochs": 3,
-  "total_flos": 112878157824000.0,
+  "max_steps": 144,
+  "num_train_epochs": 4,
+  "total_flos": 146715475968000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4d2cede8ad0eb56c1a914509bcb224f91e3e03e5504f88d252a2fa457cfb418
+oid sha256:7929aea55869137e402b1d784b7ce4ed38a6fe01ebc5666f6b57abddc466c8ff
 size 3055