huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/bring-me-the-horizon")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/10m71qwf/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/1yuqm1pv/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 Bring Me The Horizon's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2h4xcfis) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/z98a5j5r) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2h4xcfis/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/z98a5j5r/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/bring-me-the-horizon",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.672280788421631, "eval_runtime": 0.5431, "eval_samples_per_second": 75.493, "eval_steps_per_second": 11.048, "epoch": 50.0}
+{"eval_loss": 1.761746883392334, "eval_runtime": 0.4813, "eval_samples_per_second": 74.8, "eval_steps_per_second": 10.389, "epoch": 52.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db537ffe72a3ff6f07e80705e1c0c610eaa3eab6eba821179ebe2b0a97b96e66
+oid sha256:7f60e50a7e388c49df33a2cac5abe57958ed061c6525713fbbaf0e72692b8751
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a195d9dcb9ddd9fe9fc32e35077eae701ed277e42bfa38724413ed25719ce4aa
+oid sha256:0aec99ac123784c1df18282b656728c10ab998edaa76607406f49892e8c5d190
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8c7fa95bd475ba5b53141d6b6a3d95ee5f4e1f608d084240003e51a964c7b30
+oid sha256:00a416d9b660bc7b7cd01571f1569a51e5c89dfe326c7ea762cb31673f6bdd8a
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c9bdadfeacd6791e33b8118f0afae9a6b95a94da0f6e2c990e9113276a5c353
+oid sha256:9a9eb709ab35df7f70905c14ca74e38db768088a1b8d7f8fd0df5af8ff6128c9
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d34af555ef0ec293744ffd94e8e92af652827e1907b71bf5e93415c53aafec6f
+oid sha256:2bbb036f11beb59109336ec9fefde1522b1dc6499163e8831bd87d849eba417d
 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/bring-me-the-horizon", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.672280788421631,
-  "best_model_checkpoint": "output/bring-me-the-horizon/checkpoint-52",
-  "epoch": 2.0,
-  "global_step": 52,
+  "best_metric": 1.761746883392334,
+  "best_model_checkpoint": "output/bring-me-the-horizon/checkpoint-208",
+  "epoch": 8.0,
+  "global_step": 208,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -82,11 +82,245 @@
       "eval_samples_per_second": 74.866,
       "eval_steps_per_second": 10.956,
       "step": 52
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00013274211424821946,
+      "loss": 2.2735,
+      "step": 55
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.00010756924162575734,
+      "loss": 2.1555,
+      "step": 60
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 6.860000000000001e-05,
+      "loss": 2.1802,
+      "step": 65
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.9630758374242683e-05,
+      "loss": 1.978,
+      "step": 70
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.457885751780558e-06,
+      "loss": 2.4223,
+      "step": 75
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.926320195198059,
+      "eval_runtime": 0.4807,
+      "eval_samples_per_second": 74.883,
+      "eval_steps_per_second": 10.4,
+      "step": 78
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 1.9933913245728396e-06,
+      "loss": 2.1228,
+      "step": 80
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 2.310978564468141e-05,
+      "loss": 1.9814,
+      "step": 85
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 6.033118373448485e-05,
+      "loss": 2.1607,
+      "step": 90
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 0.00010048000960220248,
+      "loss": 2.1746,
+      "step": 95
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 0.00012934228335981018,
+      "loss": 1.8565,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8828926086425781,
+      "eval_runtime": 0.4815,
+      "eval_samples_per_second": 74.766,
+      "eval_steps_per_second": 10.384,
+      "step": 104
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 0.0001366998287631265,
+      "loss": 2.1214,
+      "step": 105
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 0.00011994783732453755,
+      "loss": 1.898,
+      "step": 110
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 8.501705457012643e-05,
+      "loss": 1.8994,
+      "step": 115
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 4.42741047488822e-05,
+      "loss": 1.9389,
+      "step": 120
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 1.2143306799695228e-05,
+      "loss": 1.7737,
+      "step": 125
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 1.7275,
+      "step": 130
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.8320238590240479,
+      "eval_runtime": 0.4801,
+      "eval_samples_per_second": 74.982,
+      "eval_steps_per_second": 10.414,
+      "step": 130
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 1.2143306799695106e-05,
+      "loss": 1.6759,
+      "step": 135
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 4.4274104748882125e-05,
+      "loss": 1.934,
+      "step": 140
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 8.501705457012647e-05,
+      "loss": 1.7796,
+      "step": 145
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 0.00011994783732453749,
+      "loss": 1.5568,
+      "step": 150
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 0.00013669982876312649,
+      "loss": 1.7681,
+      "step": 155
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.8055299520492554,
+      "eval_runtime": 0.4812,
+      "eval_samples_per_second": 74.816,
+      "eval_steps_per_second": 10.391,
+      "step": 156
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 0.00012934228335981018,
+      "loss": 1.5298,
+      "step": 160
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 0.00010048000960220254,
+      "loss": 1.5527,
+      "step": 165
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 6.033118373448493e-05,
+      "loss": 1.6001,
+      "step": 170
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 2.3109785644681573e-05,
+      "loss": 1.5066,
+      "step": 175
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 1.9933913245728244e-06,
+      "loss": 1.5358,
+      "step": 180
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.7874510288238525,
+      "eval_runtime": 0.4807,
+      "eval_samples_per_second": 74.894,
+      "eval_steps_per_second": 10.402,
+      "step": 182
+    },
+    {
+      "epoch": 7.12,
+      "learning_rate": 4.457885751780527e-06,
+      "loss": 1.45,
+      "step": 185
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 2.963075837424261e-05,
+      "loss": 1.4034,
+      "step": 190
+    },
+    {
+      "epoch": 7.5,
+      "learning_rate": 6.859999999999982e-05,
+      "loss": 1.3896,
+      "step": 195
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 0.00010756924162575728,
+      "loss": 1.3993,
+      "step": 200
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 0.00013274211424821943,
+      "loss": 1.4986,
+      "step": 205
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.761746883392334,
+      "eval_runtime": 0.4842,
+      "eval_samples_per_second": 74.353,
+      "eval_steps_per_second": 10.327,
+      "step": 208
     }
   ],
-  "max_steps": 1300,
-  "num_train_epochs": 50,
-  "total_flos": 52780990464000.0,
+  "max_steps": 1352,
+  "num_train_epochs": 52,
+  "total_flos": 215043342336000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:113255f50e87af9ff4033495785f9a42c4e9419fab63846e2aa6036bdea1f7dc
+oid sha256:1a61175f6b4282876536474ec422652e78dcd99a973713a9795eb16ce4347ccb
 size 2863