huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/gspd")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/17su8b0l/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/3jof0sex/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 GSPD's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/cnw6rr26) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2nxhrny4) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/cnw6rr26/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2nxhrny4/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "gspd",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -18,7 +18,9 @@
   "n_inner": null,
   "n_layer": 12,
   "n_positions": 1024,
+  "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
   "scale_attn_weights": true,
   "summary_activation": null,
   "summary_first_dropout": 0.1,
@@ -35,7 +37,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.10.0",
+  "transformers_version": "4.21.1",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 1.6143786907196045, "eval_runtime": 6.648, "eval_samples_per_second": 22.563, "eval_steps_per_second": 2.858, "epoch": 1.0}
+{"eval_loss": 1.6192371845245361, "eval_runtime": 3.877, "eval_samples_per_second": 44.106, "eval_steps_per_second": 5.674, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:903a60b019c3189ea2401d825a51588abe0295c21127eadc7f170f86b4711695
+oid sha256:6d7317d7b2d5b31cf6792c6f9b1aae151024c566ab5d2cdad309b0d8787e2d3e
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bf9330ebc61aeb2214f532e874e79d806f5773424f220a2a22e25116ae7d7ab
-size 995603825
+oid sha256:db51e790050e6606874c3cbde9a56238c405ff67191cdab7b7eecb47d32f638a
+size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:238219c7329b0b10459c22d9cd9fd9efb44970d69676335af067af5088ae373d
-size 510403817
+oid sha256:088ce82b45d839e12925669e84a76044f7f5dffaca9ddb8a53fb30425464187c
+size 510396521

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1faa9800090f58192e975d9de4af52a7d1d95073c9339bf79e7455df2e035748
+oid sha256:609e3e95ba1d71e778b2f2f711f521fee32d49f7719314a7a6f9c7574f7692ec
 size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1c2f47e2da491a315f90603f1d3a2627d77a6f03326673b420719eb45024ca9
+oid sha256:5643af0d2d309ea9f8d4e15ffbc8ebd3faca59b9cbbd8bf85f08e202e280408b
 size 623

special_tokens_map.json

@@ -1 +1,5 @@
-{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -1 +1,10 @@
-{"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"}
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1024,
+  "name_or_path": "huggingartists/gspd",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

trainer_state.json

@@ -1,144 +1,144 @@
 {
-  "best_metric": 1.6143786907196045,
-  "best_model_checkpoint": "output/gspd/checkpoint-104",
+  "best_metric": 1.6192371845245361,
+  "best_model_checkpoint": "output/gspd/checkpoint-102",
   "epoch": 1.0,
-  "global_step": 104,
+  "global_step": 102,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.05,
-      "learning_rate": 0.00013641901727743684,
-      "loss": 2.3995,
+      "learning_rate": 0.00013638815138477438,
+      "loss": 1.4771,
       "step": 5
     },
     {
       "epoch": 0.1,
-      "learning_rate": 0.00013409385144248624,
-      "loss": 2.0772,
+      "learning_rate": 0.00013397182122930294,
+      "loss": 1.6873,
       "step": 10
     },
     {
-      "epoch": 0.14,
-      "learning_rate": 0.00013027744460430822,
-      "loss": 1.9236,
+      "epoch": 0.15,
+      "learning_rate": 0.0001300082017869573,
+      "loss": 1.5944,
       "step": 15
     },
     {
-      "epoch": 0.19,
-      "learning_rate": 0.00012505669320030482,
-      "loss": 1.9836,
+      "epoch": 0.2,
+      "learning_rate": 0.0001245911081876368,
+      "loss": 1.563,
       "step": 20
     },
     {
-      "epoch": 0.24,
-      "learning_rate": 0.00011855046943587908,
-      "loss": 1.876,
+      "epoch": 0.25,
+      "learning_rate": 0.0001178487579222207,
+      "loss": 1.5602,
       "step": 25
     },
     {
       "epoch": 0.29,
-      "learning_rate": 0.00011090691466231807,
-      "loss": 1.9239,
+      "learning_rate": 0.00010994073605561706,
+      "loss": 1.5302,
       "step": 30
     },
     {
       "epoch": 0.34,
-      "learning_rate": 0.00010230006632035399,
-      "loss": 1.8628,
+      "learning_rate": 0.00010105421799895028,
+      "loss": 1.5424,
       "step": 35
     },
     {
-      "epoch": 0.38,
-      "learning_rate": 9.292589525111794e-05,
-      "loss": 1.6981,
+      "epoch": 0.39,
+      "learning_rate": 9.139953924430467e-05,
+      "loss": 1.6088,
       "step": 40
     },
     {
-      "epoch": 0.43,
-      "learning_rate": 8.299784360164853e-05,
-      "loss": 1.7632,
+      "epoch": 0.44,
+      "learning_rate": 8.120521692221671e-05,
+      "loss": 1.6842,
       "step": 45
     },
     {
-      "epoch": 0.48,
-      "learning_rate": 7.274196492316882e-05,
-      "loss": 1.8217,
+      "epoch": 0.49,
+      "learning_rate": 7.071254101695329e-05,
+      "loss": 1.5729,
       "step": 50
     },
     {
-      "epoch": 0.53,
-      "learning_rate": 6.239177711808685e-05,
-      "loss": 1.7681,
+      "epoch": 0.54,
+      "learning_rate": 6.0169863260400656e-05,
+      "loss": 1.4031,
       "step": 55
     },
     {
-      "epoch": 0.58,
-      "learning_rate": 5.218294542987356e-05,
-      "loss": 1.7394,
+      "epoch": 0.59,
+      "learning_rate": 4.9826718881055135e-05,
+      "loss": 1.4452,
       "step": 60
     },
     {
-      "epoch": 0.62,
-      "learning_rate": 4.2347916539754844e-05,
-      "loss": 1.8022,
+      "epoch": 0.64,
+      "learning_rate": 3.992792034076664e-05,
+      "loss": 1.6059,
       "step": 65
     },
     {
-      "epoch": 0.67,
-      "learning_rate": 3.311062594741274e-05,
-      "loss": 1.7416,
+      "epoch": 0.69,
+      "learning_rate": 3.0707762854909365e-05,
+      "loss": 1.4848,
       "step": 70
     },
     {
-      "epoch": 0.72,
-      "learning_rate": 2.4681399144273327e-05,
-      "loss": 1.7103,
+      "epoch": 0.74,
+      "learning_rate": 2.2384478845846205e-05,
+      "loss": 1.531,
       "step": 75
     },
     {
-      "epoch": 0.77,
-      "learning_rate": 1.725216267546246e-05,
-      "loss": 1.7463,
+      "epoch": 0.78,
+      "learning_rate": 1.515507258753902e-05,
+      "loss": 1.4834,
       "step": 80
     },
     {
-      "epoch": 0.82,
-      "learning_rate": 1.0992074130600507e-05,
-      "loss": 1.6627,
+      "epoch": 0.83,
+      "learning_rate": 9.190657300387498e-06,
+      "loss": 1.4458,
       "step": 85
     },
     {
-      "epoch": 0.87,
-      "learning_rate": 6.043670564942622e-06,
-      "loss": 1.7961,
+      "epoch": 0.88,
+      "learning_rate": 4.6324050628612214e-06,
+      "loss": 1.5666,
       "step": 90
     },
     {
-      "epoch": 0.91,
-      "learning_rate": 2.519623048140383e-06,
-      "loss": 1.7855,
+      "epoch": 0.93,
+      "learning_rate": 1.588205401691401e-06,
+      "loss": 1.5073,
       "step": 95
     },
     {
-      "epoch": 0.96,
-      "learning_rate": 5.001712368734975e-07,
-      "loss": 1.7266,
+      "epoch": 0.98,
+      "learning_rate": 1.3011164863877445e-07,
+      "loss": 1.4195,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.6143786907196045,
-      "eval_runtime": 6.5105,
-      "eval_samples_per_second": 23.04,
-      "eval_steps_per_second": 2.918,
-      "step": 104
+      "eval_loss": 1.6192371845245361,
+      "eval_runtime": 3.8612,
+      "eval_samples_per_second": 44.287,
+      "eval_steps_per_second": 5.698,
+      "step": 102
     }
   ],
-  "max_steps": 104,
-  "num_train_epochs": 1,
-  "total_flos": 108566839296000.0,
+  "max_steps": 204,
+  "num_train_epochs": 2,
+  "total_flos": 105823272960000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87567da31bf0b1dfd45bb9ff251c20e21f066cdbbeca02722787462d643334e0
-size 2671
+oid sha256:91a9ae3b4198707837d8729d85e9c1c901dba6204ce15dbd4033f413af824434
+size 3375