huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/tool")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2i0nfbt3/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/2w1h70ok/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 Tool's lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/35jhzbcz) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1zikehwi) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/35jhzbcz/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1zikehwi/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

@@ -30,8 +30,8 @@
   "task_specific_params": {
     "text-generation": {
       "do_sample": true,
-      "max_length": 200,
-      "min_length": 100,
+      "max_length": 150,
+      "min_length": 80,
       "temperature": 1.0,
       "top_p": 0.95
     }

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.619758367538452, "eval_runtime": 0.6775, "eval_samples_per_second": 19.187, "eval_steps_per_second": 2.952, "epoch": 4.0}
+{"eval_loss": 2.4025585651397705, "eval_runtime": 0.1885, "eval_samples_per_second": 79.587, "eval_steps_per_second": 10.612, "epoch": 5.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8269b8328a0bf533c21e8a9cf8abb1251769e6472a5655d46e0263fa5c36c202
+oid sha256:59dff47de0389edfe6c404dcee5ed4f8c4ee963861d19bfeb125bd68aade1e7a
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffe7e75417a67202ed0582065f1a9bd973196ecaaf6283e0353d348b8b2c9ce5
+oid sha256:ccad7917a97542e610dc3ed8f6c0eb8cb1345f9ccee69a38ffa843d74ec7ebd3
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cda16091180e330446812bef2213ce1d274c30224756be5cc5af4152ec723633
+oid sha256:b7cb352f906fb72d54412d702d0d396c7534a832d630769bbe9fca89f18763ab
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:594209d63fbcb9f59cdfbe58a2c20b27bc86dba538eb2c90c6868cbdfcae50dc
+oid sha256:ac06a0a74e2cc1058f2f6b0e902120bc773c303ad8f07aaaabab44c3ab1b8ad3
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46ac68174300d9f94d4de67167852595a5a6202d18703222cb9d7154becc9e21
+oid sha256:516bee0a3338981961397871d9fa5b0b01ff9ed881ae03d42fa50217bde1fa30
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.619758367538452,
-  "best_model_checkpoint": "output/tool/checkpoint-44",
-  "epoch": 4.0,
-  "global_step": 44,
+  "best_metric": 2.4025585651397705,
+  "best_model_checkpoint": "output/tool/checkpoint-55",
+  "epoch": 5.0,
+  "global_step": 55,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -86,11 +86,37 @@
       "eval_samples_per_second": 23.33,
       "eval_steps_per_second": 3.589,
       "step": 44
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 0.00013442121798995453,
+      "loss": 2.4054,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 5.883720209445262e-05,
+      "loss": 2.2843,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 2.4935,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.4025585651397705,
+      "eval_runtime": 0.1869,
+      "eval_samples_per_second": 80.273,
+      "eval_steps_per_second": 10.703,
+      "step": 55
     }
   ],
-  "max_steps": 44,
-  "num_train_epochs": 4,
-  "total_flos": 45987397632000.0,
+  "max_steps": 55,
+  "num_train_epochs": 5,
+  "total_flos": 57222955008000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:238f0800c1a375eef579856b16d2fb58efef9b9559fe5020f52fdd21d2d5f559
-size 3055
+oid sha256:592b1f50c346e9d470c4bc200ee6c5910a0e9902cb3de28db61f17131da5e560
+size 2991