huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/eminem")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2mu8rlhb/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/4s2qerso/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 Eminem's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1q4coblt) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1ck5hcvf) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1q4coblt/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1ck5hcvf/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "huggingartists/eminem",
+  "_name_or_path": "eminem",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -37,7 +37,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.12.3",
+  "transformers_version": "4.15.0",
   "use_cache": true,
   "vocab_size": 50257
 }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 0.6314056515693665, "eval_runtime": 29.1648, "eval_samples_per_second": 22.733, "eval_steps_per_second": 2.846, "epoch": 13.0}
+{"eval_loss": 0.4552081823348999, "eval_runtime": 28.3596, "eval_samples_per_second": 21.827, "eval_steps_per_second": 2.75, "epoch": 3.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d53934603e76bd1cc6dbaa3653dfe3ef3e8368445ee4bfb94e039bbde67a7ac6
+oid sha256:640dfd549bcdb5c64e303572f085c95c16d5ebe61a0058350dffda9fb32874d6
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a16aa18678c770ac196d53588640c9561fd9d548849b15993a4867d230b4bd96
+oid sha256:ba7722fce58191c71466f428a64d01c61d0bb9f30ab7cb137fbcff4ec53807f1
 size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:462a6a517830a28f8813a5a4aa035ae13859bf6fde9a6ed55cc4cef7924f9166
+oid sha256:613eedc4210abf9b706cb2585b2830a5899ff79ebf8f9e7495fc42c8a1eaae2b
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e00ffb84d3fc3a295652f9d6fe02a9b95ec5523997f990cc24f5bbe440e9171
+oid sha256:f7d12d6ed174599698df42e61c41e77e967775fb83fe49c3b8ec3abca5189394
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f461262713d9105e94c19b3e22fda3c3bd5c05297eb98bbd2be82f70f0da5b64
+oid sha256:1139e789b0088456b7b72a6f40cd54de1755f78a857dcab2dd97ebfb52b3f533
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.6314056515693665,
-  "best_model_checkpoint": "output/eminem/checkpoint-904",
+  "best_metric": 0.4552081823348999,
+  "best_model_checkpoint": "output/eminem/checkpoint-916",
   "epoch": 2.0,
-  "global_step": 904,
+  "global_step": 916,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1102,11 +1102,37 @@
       "eval_samples_per_second": 22.451,
       "eval_steps_per_second": 2.811,
       "step": 904
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.5530896428035615e-05,
+      "loss": 0.6893,
+      "step": 905
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 4.3329156273475095e-05,
+      "loss": 0.9131,
+      "step": 910
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.115713866225047e-05,
+      "loss": 0.8774,
+      "step": 915
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.4552081823348999,
+      "eval_runtime": 27.3413,
+      "eval_samples_per_second": 22.64,
+      "eval_steps_per_second": 2.853,
+      "step": 916
     }
   ],
-  "max_steps": 5876,
-  "num_train_epochs": 13,
-  "total_flos": 944309403648000.0,
+  "max_steps": 1374,
+  "num_train_epochs": 3,
+  "total_flos": 956198191104000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ed41b50721ec908b04ec01203c5cd320d7547c9b186f01fce593a22f969ca01
-size 2863
+oid sha256:9cc1c03bcdb4ec1c737f7dc9de1e37f89cc954f6f5a4e9df5cfc717cc7aac606
+size 2991