huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/xxxtentacion")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/wyqwtyke/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/1fgk6duz/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 XXXTENTACION's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2x3eplag) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/7g46vrtf) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2x3eplag/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/7g46vrtf/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/xxxtentacion",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -35,7 +35,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.9.2",
+  "transformers_version": "4.10.0",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 3.047952175140381, "eval_runtime": 6.9475, "eval_samples_per_second": 21.015, "eval_steps_per_second": 2.735, "epoch": 1.0}
+{"eval_loss": 2.998302936553955, "eval_runtime": 7.549, "eval_samples_per_second": 21.062, "eval_steps_per_second": 2.649, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e31086314908b688e94dea0204e120631a3c14a2a6cca0183f14233865948b8b
+oid sha256:473c6ce96484afb22b5f44b9754253982be214dde5ccfc6e4636c65d85527beb
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57fded92ce9908006c8c0c695c50b6321ba18e3a1898a7ae1ece83f466b0420a
-size 995603825
+oid sha256:42035e763018344e9c12b57960c5622bcb42c42ca8545f22d5745457bc43b7f0
+size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e27ba50977e8aa3f88b1483e1ff32e00f8e54b9b104e94626aaaf4109400399
+oid sha256:b9bc902bf8c5fd91ac6140769cbcb1ed04fda15388d328a739a52c62d0e2b7c4
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d08c08d76d72721a447ad4765dc955cf5c332927d2829e29207096ce695ec488
+oid sha256:3691a8950b6ce0a7fbcdee171960f1672959fb5e260f352ba982df89dd517c41
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5bc99df78269e2d0dbaca15016af30dea6ca5b37f97a8df9318aec4d0d9c188
+oid sha256:c0816ea5e3b1d810ba13a63e3db58b55d3131231050f4d5136f720f83ed0afbf
 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/xxxtentacion", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,156 +1,302 @@
 {
-  "best_metric": 3.047952175140381,
-  "best_model_checkpoint": "output/xxxtentacion/checkpoint-114",
-  "epoch": 1.0,
-  "global_step": 114,
+  "best_metric": 2.998302936553955,
+  "best_model_checkpoint": "output/xxxtentacion/checkpoint-226",
+  "epoch": 2.0,
+  "global_step": 226,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.04,
-      "learning_rate": 0.00013654981500613273,
-      "loss": 3.7924,
+      "learning_rate": 0.0001365382750064078,
+      "loss": 2.8729,
       "step": 5
     },
     {
       "epoch": 0.09,
-      "learning_rate": 0.00013461158482121457,
-      "loss": 3.5933,
+      "learning_rate": 0.00013456586620543126,
+      "loss": 2.8615,
       "step": 10
     },
     {
       "epoch": 0.13,
-      "learning_rate": 0.00013142205020853694,
-      "loss": 3.4387,
+      "learning_rate": 0.0001313208258478339,
+      "loss": 2.8047,
       "step": 15
     },
     {
       "epoch": 0.18,
-      "learning_rate": 0.00012704167144630924,
-      "loss": 3.3335,
+      "learning_rate": 0.00012686575814089965,
+      "loss": 2.9533,
       "step": 20
     },
     {
       "epoch": 0.22,
-      "learning_rate": 0.00012155348225285961,
-      "loss": 3.3589,
+      "learning_rate": 0.00012128661147092064,
+      "loss": 2.8864,
       "step": 25
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 0.00011506151581352585,
-      "loss": 3.3227,
+      "epoch": 0.27,
+      "learning_rate": 0.00011469102026339097,
+      "loss": 2.5404,
       "step": 30
     },
     {
       "epoch": 0.31,
-      "learning_rate": 0.00010768883274520426,
-      "loss": 3.6526,
+      "learning_rate": 0.00010720622847019214,
+      "loss": 3.0059,
       "step": 35
     },
     {
       "epoch": 0.35,
-      "learning_rate": 9.957518838012702e-05,
-      "loss": 3.2754,
+      "learning_rate": 9.897663474442597e-05,
+      "loss": 2.827,
       "step": 40
     },
     {
-      "epoch": 0.39,
-      "learning_rate": 9.08743835874413e-05,
-      "loss": 3.2572,
+      "epoch": 0.4,
+      "learning_rate": 9.01610066620532e-05,
+      "loss": 2.804,
       "step": 45
     },
     {
       "epoch": 0.44,
-      "learning_rate": 8.175134934996847e-05,
-      "loss": 3.1639,
+      "learning_rate": 8.092941773433672e-05,
+      "loss": 2.8347,
       "step": 50
     },
     {
-      "epoch": 0.48,
-      "learning_rate": 7.237902036041238e-05,
-      "loss": 3.343,
+      "epoch": 0.49,
+      "learning_rate": 7.14599663030852e-05,
+      "loss": 2.7375,
       "step": 55
     },
     {
       "epoch": 0.53,
-      "learning_rate": 6.293505690059801e-05,
-      "loss": 3.4637,
+      "learning_rate": 6.193533961866806e-05,
+      "loss": 2.5601,
       "step": 60
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 5.359847714324262e-05,
-      "loss": 3.1047,
+      "epoch": 0.58,
+      "learning_rate": 5.253928938754926e-05,
+      "loss": 2.8287,
       "step": 65
     },
     {
-      "epoch": 0.61,
-      "learning_rate": 4.4546263713777056e-05,
-      "loss": 3.0869,
+      "epoch": 0.62,
+      "learning_rate": 4.345308678404285e-05,
+      "loss": 2.7017,
       "step": 70
     },
     {
       "epoch": 0.66,
-      "learning_rate": 3.5950008837656755e-05,
-      "loss": 3.3149,
+      "learning_rate": 3.4852025317174834e-05,
+      "loss": 2.7319,
       "step": 75
     },
     {
-      "epoch": 0.7,
-      "learning_rate": 2.7972661667182378e-05,
-      "loss": 3.107,
+      "epoch": 0.71,
+      "learning_rate": 2.6902039020299325e-05,
+      "loss": 2.7863,
       "step": 80
     },
     {
       "epoch": 0.75,
-      "learning_rate": 2.0765439444951916e-05,
-      "loss": 2.9627,
+      "learning_rate": 1.97565012062839e-05,
+      "loss": 2.7227,
       "step": 85
     },
     {
-      "epoch": 0.79,
-      "learning_rate": 1.4464961055407408e-05,
-      "loss": 3.0099,
+      "epoch": 0.8,
+      "learning_rate": 1.3553265547553997e-05,
+      "loss": 2.4648,
       "step": 90
     },
     {
-      "epoch": 0.83,
-      "learning_rate": 9.190657300387505e-06,
-      "loss": 3.5708,
+      "epoch": 0.84,
+      "learning_rate": 8.412006565291007e-06,
+      "loss": 2.4789,
       "step": 95
     },
     {
       "epoch": 0.88,
-      "learning_rate": 5.042506989064599e-06,
-      "loss": 3.0849,
+      "learning_rate": 4.431910835795193e-06,
+      "loss": 2.6082,
       "step": 100
     },
     {
-      "epoch": 0.92,
-      "learning_rate": 2.0991417565619363e-06,
-      "loss": 3.1334,
+      "epoch": 0.93,
+      "learning_rate": 1.6897634558914594e-06,
+      "loss": 2.8574,
       "step": 105
     },
     {
-      "epoch": 0.96,
-      "learning_rate": 4.1635553598866533e-07,
-      "loss": 3.2816,
+      "epoch": 0.97,
+      "learning_rate": 2.384666838129088e-07,
+      "loss": 2.5917,
       "step": 110
     },
     {
       "epoch": 1.0,
-      "eval_loss": 3.047952175140381,
-      "eval_runtime": 6.9391,
-      "eval_samples_per_second": 21.04,
-      "eval_steps_per_second": 2.738,
-      "step": 114
+      "eval_loss": 3.0134646892547607,
+      "eval_runtime": 7.5402,
+      "eval_samples_per_second": 21.087,
+      "eval_steps_per_second": 2.652,
+      "step": 113
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0601933434684325e-07,
+      "loss": 2.4012,
+      "step": 115
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2949766180319454e-06,
+      "loss": 2.5116,
+      "step": 120
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.7824008453368197e-06,
+      "loss": 2.682,
+      "step": 125
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 7.5203039468480955e-06,
+      "loss": 2.6505,
+      "step": 130
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.2436573272248243e-05,
+      "loss": 2.5776,
+      "step": 135
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.8436362807314313e-05,
+      "loss": 2.631,
+      "step": 140
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.540392296919934e-05,
+      "loss": 2.4606,
+      "step": 145
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.320483367908481e-05,
+      "loss": 2.3774,
+      "step": 150
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.1688597631154055e-05,
+      "loss": 2.5596,
+      "step": 155
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.06915437278321e-05,
+      "loss": 2.5551,
+      "step": 160
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 6.003998466742405e-05,
+      "loss": 2.4846,
+      "step": 165
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 6.955356776710644e-05,
+      "loss": 2.5243,
+      "step": 170
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 7.904875437627624e-05,
+      "loss": 2.5396,
+      "step": 175
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 8.834236075446375e-05,
+      "loss": 2.4629,
+      "step": 180
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.725509210222169e-05,
+      "loss": 2.5566,
+      "step": 185
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00010561500156549765,
+      "loss": 2.5873,
+      "step": 190
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.00011326080748143412,
+      "loss": 2.5575,
+      "step": 195
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00012004500486837714,
+      "loss": 2.3709,
+      "step": 200
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.0001258367111323791,
+      "loss": 2.6089,
+      "step": 205
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.0001305241910900454,
+      "loss": 2.5663,
+      "step": 210
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00013401701259429296,
+      "loss": 2.6963,
+      "step": 215
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00013624779117609787,
+      "loss": 2.6391,
+      "step": 220
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.0001371734900441264,
+      "loss": 2.7449,
+      "step": 225
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.998302936553955,
+      "eval_runtime": 7.5393,
+      "eval_samples_per_second": 21.089,
+      "eval_steps_per_second": 2.653,
+      "step": 226
     }
   ],
-  "max_steps": 114,
-  "num_train_epochs": 1,
-  "total_flos": 118887874560000.0,
+  "max_steps": 226,
+  "num_train_epochs": 2,
+  "total_flos": 234378952704000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e74c5392a6e0e5419a3c8e54cc2a6e2abd8a96dda1e8be3831c74bd250130fe2
+oid sha256:82f2723ca1e95981d792cb45af7783a851ea2dee4eed340a95ef82a8fb6a5a29
 size 2671