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/2hhippgr/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/tusv3cma/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/actqj6q5) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/kcn90ud5) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/actqj6q5/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/kcn90ud5/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -8,7 +8,6 @@
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
-  "gradient_checkpointing": false,
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 0.17592251300811768, "eval_runtime": 14.7701, "eval_samples_per_second": 43.669, "eval_steps_per_second": 5.484, "epoch": 3.0}
+{"eval_loss": 0.15162613987922668, "eval_runtime": 14.5602, "eval_samples_per_second": 44.23, "eval_steps_per_second": 5.563, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8867b432c877cd9596b4a8b7fad6cfb7e38ba718fa221a48c5ebbb4fdde6e544
+oid sha256:f7761ecce8eb19c94742f6c9fd9bee473e591dcc45cbcb2ee94e93dee113edf7
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f716e95e3dbbcbe168dc20c1c3b7e31df929091c7313362e852e9a7075e8397c
+oid sha256:4476e576f6e7c72bcefb1035b67b146bab7f5ece7faedb85e92bffb11d4796ce
 size 995604017

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76514c0aa443b573a0a64185ca920e8d304a43122010a4fbb1e360063fd56c96
+oid sha256:5516fb13956d24eaaff944923fb65d6b9148d14860382a06da5c8132587cd3dc
 size 510396521

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d7e2b7a1002f45683fa60c6f3807d23bfc69c51e189e8618fae3d847cb1f5a7
+oid sha256:27912df59e20e4b2d528762eeb52cf6b416a98f296d07fe6caab6324eee7cd93
 size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02691acbe003cbe61f74603e7458e8f92726bc52ceb9a64525eebc387d776947
+oid sha256:7dd0d4fbaf273821538ccb639a7a35444f70ec1b31c607e5a8601956d595803c
 size 623

tokenizer_config.json

@@ -3,7 +3,7 @@
   "bos_token": "<|endoftext|>",
   "eos_token": "<|endoftext|>",
   "model_max_length": 1024,
-  "name_or_path": "huggingartists/eminem",
+  "name_or_path": "gpt2",
   "special_tokens_map_file": null,
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<|endoftext|>"

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.17592251300811768,
-  "best_model_checkpoint": "output/eminem/checkpoint-454",
+  "best_metric": 0.15162613987922668,
+  "best_model_checkpoint": "output/eminem/checkpoint-455",
   "epoch": 1.0,
-  "global_step": 454,
+  "global_step": 455,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -554,11 +554,25 @@
       "eval_samples_per_second": 43.502,
       "eval_steps_per_second": 5.463,
       "step": 454
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.181858176201967e-07,
+      "loss": 0.3187,
+      "step": 455
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.15162613987922668,
+      "eval_runtime": 13.378,
+      "eval_samples_per_second": 48.139,
+      "eval_steps_per_second": 6.055,
+      "step": 455
     }
   ],
-  "max_steps": 1362,
-  "num_train_epochs": 3,
-  "total_flos": 474506330112000.0,
+  "max_steps": 910,
+  "num_train_epochs": 2,
+  "total_flos": 474636976128000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15d65f1ac2b1e539f6ce4f76e7d3bd07cdd75d9a6400c077f2ebc2c9e23d2c6d
+oid sha256:8430f2d6c4d4f936bbbf8c552b5578f9269d0dd84df240e93b07571cd47afa8e
 size 3311