huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/pharaoh")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2j2p1ulk/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/3kao1xf5/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 PHARAOH's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1dgmwwm3) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/xsxegmjq) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1dgmwwm3/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/xsxegmjq/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/pharaoh",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -35,7 +35,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.10.0",
+  "transformers_version": "4.11.2",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 1.699326515197754, "eval_runtime": 10.4841, "eval_samples_per_second": 22.606, "eval_steps_per_second": 2.861, "epoch": 1.0}
+{"eval_loss": 1.7224986553192139, "eval_runtime": 22.1306, "eval_samples_per_second": 22.548, "eval_steps_per_second": 2.847, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26cbcf0420766236c89194181a5733b514b4c264cdea2f09855e9306bfd740c7
+oid sha256:9a2b4c5319db04b9872d2055b5b79d1b4807f1c3915b4626047165c4d8e7d4ec
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe1daefc1a60e30524b28874d4be522c1e80a4766d1a84d1d30ff9599aabf714
-size 995603825
+oid sha256:26bbacfce71425f128a87fdb7029463499be8bc995eb14eb839740e92b47317a
+size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f77e516fe0382ad6598763db798fe469510208becc1e25a055700e3eee0fb6f
+oid sha256:bff62d0348d173b28829e07fec717386e9eef99e506d12c023aa35c55239ae7d
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f77f96ae7ce45ce31ca4af2726b72623ac2be13cdb55ba1d562b78afb740e564
-size 14503
+oid sha256:9da0f194b462662f09bef1ea47010e6352ae99c4cb094909f982d2deeb47ae8b
+size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48b660f4750a1868cd1a9877f5c933f8f45093566a7accb3c6c62e58f34d678f
+oid sha256:3ed819f11a0f4645df58a01f50e8c26eb756310ea7f77b118bb1ec57797118f8
 size 623

tokenizer_config.json

@@ -1 +1 @@
-{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/pharaoh", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,264 +1,222 @@
 {
-  "best_metric": 1.699326515197754,
-  "best_model_checkpoint": "output/pharaoh/checkpoint-200",
+  "best_metric": 1.7224986553192139,
+  "best_model_checkpoint": "output/pharaoh/checkpoint-167",
   "epoch": 1.0,
-  "global_step": 200,
+  "global_step": 167,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.03,
-      "learning_rate": 0.0001369885290940926,
-      "loss": 2.5312,
+      "learning_rate": 0.00013689676402972123,
+      "loss": 1.6111,
       "step": 5
     },
     {
-      "epoch": 0.05,
-      "learning_rate": 0.00013635542016482644,
-      "loss": 2.1558,
+      "epoch": 0.06,
+      "learning_rate": 0.0001359897369367762,
+      "loss": 1.6407,
       "step": 10
     },
     {
-      "epoch": 0.07,
-      "learning_rate": 0.0001353045765392806,
-      "loss": 2.1677,
+      "epoch": 0.09,
+      "learning_rate": 0.00013448693747453513,
+      "loss": 1.7296,
       "step": 15
     },
     {
-      "epoch": 0.1,
-      "learning_rate": 0.00013384247701784754,
-      "loss": 2.0874,
+      "epoch": 0.12,
+      "learning_rate": 0.0001324016514404635,
+      "loss": 1.6351,
       "step": 20
     },
     {
-      "epoch": 0.12,
-      "learning_rate": 0.00013197813593027427,
-      "loss": 2.0224,
+      "epoch": 0.15,
+      "learning_rate": 0.00012975231422038792,
+      "loss": 1.6431,
       "step": 25
     },
     {
-      "epoch": 0.15,
-      "learning_rate": 0.00012972304755932203,
-      "loss": 1.9402,
+      "epoch": 0.18,
+      "learning_rate": 0.0001265623478068018,
+      "loss": 1.5934,
       "step": 30
     },
     {
-      "epoch": 0.17,
-      "learning_rate": 0.00012709111527469072,
-      "loss": 2.113,
+      "epoch": 0.21,
+      "learning_rate": 0.00012285995373208355,
+      "loss": 1.6383,
       "step": 35
     },
     {
-      "epoch": 0.2,
-      "learning_rate": 0.0001240985658141214,
-      "loss": 2.0482,
+      "epoch": 0.24,
+      "learning_rate": 0.00011867786374724195,
+      "loss": 1.688,
       "step": 40
     },
     {
-      "epoch": 0.23,
-      "learning_rate": 0.00012076384924016213,
-      "loss": 1.843,
+      "epoch": 0.27,
+      "learning_rate": 0.00011405305045036307,
+      "loss": 1.7355,
       "step": 45
     },
     {
-      "epoch": 0.25,
-      "learning_rate": 0.00011710752518939715,
-      "loss": 2.0109,
+      "epoch": 0.3,
+      "learning_rate": 0.00010902640042300463,
+      "loss": 1.6672,
       "step": 50
     },
     {
-      "epoch": 0.28,
-      "learning_rate": 0.00011315213611545058,
-      "loss": 1.9608,
+      "epoch": 0.33,
+      "learning_rate": 0.00010364235276424244,
+      "loss": 1.6835,
       "step": 55
     },
     {
-      "epoch": 0.3,
-      "learning_rate": 0.00010892206830726365,
-      "loss": 2.053,
+      "epoch": 0.36,
+      "learning_rate": 9.794850621797833e-05,
+      "loss": 1.5821,
       "step": 60
     },
     {
-      "epoch": 0.33,
-      "learning_rate": 0.00010444340153951408,
-      "loss": 1.9682,
+      "epoch": 0.39,
+      "learning_rate": 9.199519836678021e-05,
+      "loss": 1.701,
       "step": 65
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 9.974374828213291e-05,
-      "loss": 1.9954,
+      "epoch": 0.42,
+      "learning_rate": 8.583506061247361e-05,
+      "loss": 1.6752,
       "step": 70
     },
     {
-      "epoch": 0.38,
-      "learning_rate": 9.485208346024516e-05,
-      "loss": 1.9485,
+      "epoch": 0.45,
+      "learning_rate": 7.95225528777695e-05,
+      "loss": 1.6845,
       "step": 75
     },
     {
-      "epoch": 0.4,
-      "learning_rate": 8.97985658141214e-05,
-      "loss": 1.8621,
+      "epoch": 0.48,
+      "learning_rate": 7.311348214248791e-05,
+      "loss": 1.6935,
       "step": 80
     },
     {
-      "epoch": 0.42,
-      "learning_rate": 8.461435196051512e-05,
-      "loss": 1.9375,
+      "epoch": 0.51,
+      "learning_rate": 6.666450907085589e-05,
+      "loss": 1.7025,
       "step": 85
     },
     {
-      "epoch": 0.45,
-      "learning_rate": 7.933140430175984e-05,
-      "loss": 1.8682,
+      "epoch": 0.54,
+      "learning_rate": 6.023264709163982e-05,
+      "loss": 1.6483,
       "step": 90
     },
     {
-      "epoch": 0.47,
-      "learning_rate": 7.398229396693017e-05,
-      "loss": 1.9141,
+      "epoch": 0.57,
+      "learning_rate": 5.387475835959079e-05,
+      "loss": 1.5692,
       "step": 95
     },
     {
-      "epoch": 0.5,
-      "learning_rate": 6.86e-05,
-      "loss": 1.8811,
+      "epoch": 0.6,
+      "learning_rate": 4.7647051054259744e-05,
+      "loss": 1.6434,
       "step": 100
     },
     {
-      "epoch": 0.53,
-      "learning_rate": 6.321770603306985e-05,
-      "loss": 1.9623,
+      "epoch": 0.63,
+      "learning_rate": 4.1604582460408433e-05,
+      "loss": 1.661,
       "step": 105
     },
     {
-      "epoch": 0.55,
-      "learning_rate": 5.786859569824015e-05,
-      "loss": 1.8849,
+      "epoch": 0.66,
+      "learning_rate": 3.580077222313727e-05,
+      "loss": 1.6785,
       "step": 110
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 5.258564803948491e-05,
-      "loss": 1.891,
+      "epoch": 0.69,
+      "learning_rate": 3.0286930080889863e-05,
+      "loss": 1.6299,
       "step": 115
     },
     {
-      "epoch": 0.6,
-      "learning_rate": 4.740143418587861e-05,
-      "loss": 1.7659,
+      "epoch": 0.72,
+      "learning_rate": 2.511180225150775e-05,
+      "loss": 1.6296,
       "step": 120
     },
     {
-      "epoch": 0.62,
-      "learning_rate": 4.2347916539754844e-05,
-      "loss": 1.8337,
+      "epoch": 0.75,
+      "learning_rate": 2.0321140481598987e-05,
+      "loss": 1.6131,
       "step": 125
     },
     {
-      "epoch": 0.65,
-      "learning_rate": 3.745625171786709e-05,
-      "loss": 1.8855,
+      "epoch": 0.78,
+      "learning_rate": 1.595729756912515e-05,
+      "loss": 1.6039,
       "step": 130
     },
     {
-      "epoch": 0.68,
-      "learning_rate": 3.275659846048591e-05,
-      "loss": 1.8395,
+      "epoch": 0.81,
+      "learning_rate": 1.2058852935068877e-05,
+      "loss": 1.696,
       "step": 135
     },
     {
-      "epoch": 0.7,
-      "learning_rate": 2.8277931692736352e-05,
-      "loss": 1.8782,
+      "epoch": 0.84,
+      "learning_rate": 8.660271554391817e-06,
+      "loss": 1.6846,
       "step": 140
     },
     {
-      "epoch": 0.72,
-      "learning_rate": 2.4047863884549408e-05,
-      "loss": 1.8944,
+      "epoch": 0.87,
+      "learning_rate": 5.791599261570497e-06,
+      "loss": 1.6253,
       "step": 145
     },
     {
-      "epoch": 0.75,
-      "learning_rate": 2.0092474810602843e-05,
-      "loss": 1.913,
+      "epoch": 0.9,
+      "learning_rate": 3.4781971244218827e-06,
+      "loss": 1.5759,
       "step": 150
     },
     {
-      "epoch": 0.78,
-      "learning_rate": 1.6436150759837878e-05,
-      "loss": 1.7538,
+      "epoch": 0.93,
+      "learning_rate": 1.7405172345409156e-06,
+      "loss": 1.628,
       "step": 155
     },
     {
-      "epoch": 0.8,
-      "learning_rate": 1.3101434185878613e-05,
-      "loss": 1.9117,
+      "epoch": 0.96,
+      "learning_rate": 5.939218965176846e-07,
+      "loss": 1.6359,
       "step": 160
     },
     {
-      "epoch": 0.82,
-      "learning_rate": 1.0108884725309277e-05,
-      "loss": 1.8817,
+      "epoch": 0.99,
+      "learning_rate": 4.8547814429227996e-08,
+      "loss": 1.614,
       "step": 165
     },
-    {
-      "epoch": 0.85,
-      "learning_rate": 7.476952440677969e-06,
-      "loss": 1.8265,
-      "step": 170
-    },
-    {
-      "epoch": 0.88,
-      "learning_rate": 5.22186406972573e-06,
-      "loss": 1.7538,
-      "step": 175
-    },
-    {
-      "epoch": 0.9,
-      "learning_rate": 3.357522982152468e-06,
-      "loss": 1.7725,
-      "step": 180
-    },
-    {
-      "epoch": 0.93,
-      "learning_rate": 1.8954234607193881e-06,
-      "loss": 1.8434,
-      "step": 185
-    },
-    {
-      "epoch": 0.95,
-      "learning_rate": 8.445798351735566e-07,
-      "loss": 1.7535,
-      "step": 190
-    },
-    {
-      "epoch": 0.97,
-      "learning_rate": 2.1147090590742169e-07,
-      "loss": 1.8311,
-      "step": 195
-    },
-    {
-      "epoch": 1.0,
-      "learning_rate": 0.0,
-      "loss": 1.862,
-      "step": 200
-    },
     {
       "epoch": 1.0,
-      "eval_loss": 1.699326515197754,
-      "eval_runtime": 10.3078,
-      "eval_samples_per_second": 22.992,
-      "eval_steps_per_second": 2.91,
-      "step": 200
+      "eval_loss": 1.7224986553192139,
+      "eval_runtime": 22.2232,
+      "eval_samples_per_second": 22.454,
+      "eval_steps_per_second": 2.835,
+      "step": 167
     }
   ],
-  "max_steps": 200,
-  "num_train_epochs": 1,
-  "total_flos": 208772333568000.0,
+  "max_steps": 334,
+  "num_train_epochs": 2,
+  "total_flos": 174412431360000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd6418f8aaf396178babd67aee6ff50924acc2cdcb8097797e754c240edd3780
-size 2671
+oid sha256:ce0544ae26ed18e0692923459bb51b32637aeec0f8caa1502a1b6bf0910ecd6d
+size 2863