huggingartists

README.md

@@ -45,15 +45,15 @@ from datasets import load_dataset
 dataset = load_dataset("huggingartists/ot-rus")
 ```
 
-[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/dh75fodw/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/35byet4r/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 O.T (RUS)'s lyrics.
 
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2z8je9ki) for full transparency and reproducibility.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2p2tawej) for full transparency and reproducibility.
 
-At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2z8je9ki/artifacts) is logged and versioned.
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2p2tawej/artifacts) is logged and versioned.
 
 ## How to use
 

config.json

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

evaluation.txt

@@ -1 +1 @@
-{"eval_loss": 2.1647870540618896, "eval_runtime": 5.6073, "eval_samples_per_second": 22.471, "eval_steps_per_second": 2.853, "epoch": 1.0}
+{"eval_loss": 2.1074936389923096, "eval_runtime": 5.5699, "eval_samples_per_second": 22.263, "eval_steps_per_second": 2.873, "epoch": 2.0}

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7280534ebae80cbd7a0ff181370ac219ef1d383b6a75ec3ace45003f7fa3b4c2
+oid sha256:07011c203546934b6bdabaeae788ab6a69d4e88dfb66510ddc805a805f61bc05
 size 497764120

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80aa3d46df7706d73aab712bda5577adf024661077c53095998751be4b4bb475
+oid sha256:14a546279a0f707235c018ab8978fcf6b535a23f50333ef00dc4c7776bfdd3b3
 size 995603825

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e920d39caf1d90c47efebeb7622ea08b9d909896b23eb32a2e43d029e6277013
+oid sha256:71ee1123b7c5a177aaa470ceeb18d3f4774881c9c284c22b3f07eb4b7fcd7595
 size 510403817

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6d05b9438e6f0b829d1a0fd700693afb37fc20834900c9a611d951a29b89e0e
-size 14567
+oid sha256:dd50da534a44c60e56efa338fe7a8a2934c906739bbceef2efb60aba8acaa3be
+size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d42e804a9d8c3eb6fb374f2bef948d45b343d1713752796e2fab2e4688dc861b
+oid sha256:5c4614765a0c82684c77b1a0e60e0bd2f59801705437e85879cf2668b5350536
 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/ot-rus", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 2.1647870540618896,
-  "best_model_checkpoint": "output/ot-rus/checkpoint-93",
-  "epoch": 1.0,
-  "global_step": 93,
+  "best_metric": 2.1074936389923096,
+  "best_model_checkpoint": "output/ot-rus/checkpoint-188",
+  "epoch": 2.0,
+  "global_step": 188,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -122,11 +122,141 @@
       "eval_samples_per_second": 23.177,
       "eval_steps_per_second": 2.943,
       "step": 93
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.152095079421997,
+      "eval_runtime": 5.3828,
+      "eval_samples_per_second": 23.036,
+      "eval_steps_per_second": 2.972,
+      "step": 94
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.830872810453363e-08,
+      "loss": 2.1472,
+      "step": 95
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.3746270344901413e-06,
+      "loss": 2.2033,
+      "step": 100
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.58381008254605e-06,
+      "loss": 2.1682,
+      "step": 105
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.576451662754438e-06,
+      "loss": 2.1479,
+      "step": 110
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.6213459316852997e-05,
+      "loss": 2.2949,
+      "step": 115
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.4309929383066146e-05,
+      "loss": 2.2226,
+      "step": 120
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.3640298318194444e-05,
+      "loss": 2.1933,
+      "step": 125
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.3944626783346644e-05,
+      "loss": 2.1917,
+      "step": 130
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 5.493584142187686e-05,
+      "loss": 2.2443,
+      "step": 135
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 6.630773257727353e-05,
+      "loss": 2.2132,
+      "step": 140
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 7.774348513864122e-05,
+      "loss": 2.2304,
+      "step": 145
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.892450484875447e-05,
+      "loss": 2.0878,
+      "step": 150
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.953929417822461e-05,
+      "loss": 2.2156,
+      "step": 155
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00010929213048843373,
+      "loss": 2.165,
+      "step": 160
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00011791130471402592,
+      "loss": 2.0622,
+      "step": 165
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00012515669103944476,
+      "loss": 2.2076,
+      "step": 170
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00013082643668217578,
+      "loss": 2.2435,
+      "step": 175
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00013476258540873022,
+      "loss": 2.1772,
+      "step": 180
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00013685547811507137,
+      "loss": 2.1434,
+      "step": 185
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.1074936389923096,
+      "eval_runtime": 5.358,
+      "eval_samples_per_second": 23.143,
+      "eval_steps_per_second": 2.986,
+      "step": 188
     }
   ],
-  "max_steps": 93,
-  "num_train_epochs": 1,
-  "total_flos": 97069989888000.0,
+  "max_steps": 188,
+  "num_train_epochs": 2,
+  "total_flos": 194793209856000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7356065e64a7ed2f99760f766f2d6d5548e73090f96e18983074ac05700b0e4
+oid sha256:b11e3b4f7a9410936e65efe533a2197fa464fb8c605563f3fdb5b97e67df2e0d
 size 2671