huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/aaron-watson")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2hy5w75w/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/204aifr7/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 Aaron Watson's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2h0fs8qq) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/20vl7rt5) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2h0fs8qq/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/20vl7rt5/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "gpt2",
+  "_name_or_path": "huggingartists/aaron-watson",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.996838331222534, "eval_runtime": 1.5779, "eval_samples_per_second": 21.547, "eval_steps_per_second": 3.169, "epoch": 2.0}
+{"eval_loss": 2.9697048664093018, "eval_runtime": 1.8075, "eval_samples_per_second": 20.47, "eval_steps_per_second": 2.766, "epoch": 5.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bed746cf4a023760b7e67a6ff2a80d140a2f44937ce586a9dad8d46e04a4b32c
+oid sha256:75c4ed6e042b687530615246e336ead94b468f8bffd1f840eeb3535c66e14912
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4dd6b9dedd4d3504b8efcee5e7b4135aa72ece84eb0a07df5d59820f745486db
+oid sha256:af00edd8315e8240fc11cc4bbe4fd8812f8e9944d63e1e804ed4135ef7995423
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad5b45a12c5b8b9ea6e74cd1f0e45fefae3e6e3e41e981df0e65db608dbd91a4
+oid sha256:4ae3248a48d48f36b9ecb728e6afed45c93ad66bd3f5adf5a346cb1e380479b5
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae45c233b14eda3cf7879d31899e37130a0d6bf44f948bbd5db7707ce01455fa
+oid sha256:e6e12081284488e0302f0f15cc7b9e867b93ab823a7f5b31fa14b05b508db300
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c40e0b165562c00350699054d9d108e03d2359770326bafc9d90a379a5ec4c14
+oid sha256:1d625cddcadf735c6fc48a4f3f1d506c16673a1b6749611364d76e3a20b91fce
 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/aaron-watson", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.996838331222534,
-  "best_model_checkpoint": "output/aaron-watson/checkpoint-50",
-  "epoch": 2.0,
-  "global_step": 50,
+  "best_metric": 2.9697048664093018,
+  "best_model_checkpoint": "output/aaron-watson/checkpoint-75",
+  "epoch": 3.0,
+  "global_step": 75,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -82,11 +82,49 @@
       "eval_samples_per_second": 21.512,
       "eval_steps_per_second": 3.163,
       "step": 50
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 3.0342,
+      "step": 55
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 8.979856581412141e-05,
+      "loss": 2.8495,
+      "step": 60
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.7401434185878625e-05,
+      "loss": 2.895,
+      "step": 65
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.310143418587862e-05,
+      "loss": 2.8209,
+      "step": 70
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 2.7285,
+      "step": 75
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.9697048664093018,
+      "eval_runtime": 1.7446,
+      "eval_samples_per_second": 21.208,
+      "eval_steps_per_second": 2.866,
+      "step": 75
     }
   ],
-  "max_steps": 50,
-  "num_train_epochs": 2,
-  "total_flos": 51474530304000.0,
+  "max_steps": 125,
+  "num_train_epochs": 5,
+  "total_flos": 76689211392000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc58aa350caebd0c275c3294305ba485a147a6157454835c10b9e04bb0858638
+oid sha256:66237f886a59d41c0e749b69ee85bc3f1c09d6f131ed3be21ee7e5ab55356657
 size 2671