joelniklaus commited on Feb 10, 2023

Commit

b4d1884

•

1 Parent(s): ff34cd6

Training in progress, step 50000

Browse files

Files changed (26) hide show

.gitignore +1 -0
config.json +27 -0
last-checkpoint/config.json +27 -0
last-checkpoint/optimizer.pt +3 -0
last-checkpoint/pytorch_model.bin +3 -0
last-checkpoint/rng_state_0.pth +3 -0
last-checkpoint/rng_state_1.pth +3 -0
last-checkpoint/rng_state_2.pth +3 -0
last-checkpoint/rng_state_3.pth +3 -0
last-checkpoint/rng_state_4.pth +3 -0
last-checkpoint/rng_state_5.pth +3 -0
last-checkpoint/rng_state_6.pth +3 -0
last-checkpoint/rng_state_7.pth +3 -0
last-checkpoint/scheduler.pt +3 -0
last-checkpoint/special_tokens_map.json +9 -0
last-checkpoint/tokenizer.json +0 -0
last-checkpoint/tokenizer_config.json +13 -0
last-checkpoint/trainer_state.json +324 -0
last-checkpoint/training_args.bin +3 -0
pytorch_model.bin +3 -0
runs/Feb09_18-26-34_t1v-n-07f0074d-w-0/1675967214.712931/events.out.tfevents.1675967214.t1v-n-07f0074d-w-0.3787133.1 +3 -0
runs/Feb09_18-26-34_t1v-n-07f0074d-w-0/events.out.tfevents.1675967214.t1v-n-07f0074d-w-0.3787133.0 +3 -0
special_tokens_map.json +9 -0
tokenizer.json +0 -0
tokenizer_config.json +13 -0
training_args.bin +3 -0

.gitignore ADDED Viewed

	@@ -0,0 +1 @@


1	+ checkpoint-*/

config.json ADDED Viewed

	@@ -0,0 +1,27 @@

+{
+  "_name_or_path": "data/plms/legal-lithuanian-roberta-base",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 1,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

last-checkpoint/config.json ADDED Viewed

	@@ -0,0 +1,27 @@

+{
+  "_name_or_path": "data/plms/legal-lithuanian-roberta-base",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 1,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

last-checkpoint/optimizer.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a23c49d487e0019dc985d75287fb7093ae6d9358b291f443ffc6b3d3ac01f60a
+size 885324633

last-checkpoint/pytorch_model.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3ecccd9eb602c427ec07b2e2e082a8f8a5cac7f4aecb7f1cc99d1c0fc82cd41d
+size 442675755

last-checkpoint/rng_state_0.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_1.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_2.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_3.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_4.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_5.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_6.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/rng_state_7.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e7570246e1f2378a9fa9a7220e9a0bba4d7a3e246b3cad14f686250de7cd023
+size 13611

last-checkpoint/scheduler.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4eda5fed79fbfae57d1a589b135999ab70f7c329a9938c428058fd62dc77c658
+size 623

last-checkpoint/special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

last-checkpoint/tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

last-checkpoint/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,13 @@

+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "name_or_path": "data/plms/legal-lithuanian-roberta-base",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "special_tokens_map_file": "data/plms/legal-lithuanian-roberta-base_32k/special_tokens_map.json",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

last-checkpoint/trainer_state.json ADDED Viewed

	@@ -0,0 +1,324 @@

+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 23.00942,
+  "global_step": 50000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 1e-05,
+      "loss": 6.365,
+      "step": 1000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2e-05,
+      "loss": 4.54,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3e-05,
+      "loss": 3.9405,
+      "step": 3000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4e-05,
+      "loss": 3.6444,
+      "step": 4000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5e-05,
+      "loss": 3.1408,
+      "step": 5000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6e-05,
+      "loss": 2.3019,
+      "step": 6000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7e-05,
+      "loss": 1.802,
+      "step": 7000
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 8e-05,
+      "loss": 1.6724,
+      "step": 8000
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 9e-05,
+      "loss": 1.5194,
+      "step": 9000
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 0.0001,
+      "loss": 1.4654,
+      "step": 10000
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 9.999316524962345e-05,
+      "loss": 1.4209,
+      "step": 11000
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 9.997266286704631e-05,
+      "loss": 1.3222,
+      "step": 12000
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 9.993849845741524e-05,
+      "loss": 1.3633,
+      "step": 13000
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 9.989068136093873e-05,
+      "loss": 1.2351,
+      "step": 14000
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 9.98292246503335e-05,
+      "loss": 1.3078,
+      "step": 15000
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 9.975414512725057e-05,
+      "loss": 1.1871,
+      "step": 16000
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 9.966546331768191e-05,
+      "loss": 1.2577,
+      "step": 17000
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 9.956320346634876e-05,
+      "loss": 1.1666,
+      "step": 18000
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 9.944739353007344e-05,
+      "loss": 1.2102,
+      "step": 19000
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 9.931806517013612e-05,
+      "loss": 1.1394,
+      "step": 20000
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 9.917525374361912e-05,
+      "loss": 1.1911,
+      "step": 21000
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 9.901899829374047e-05,
+      "loss": 1.1113,
+      "step": 22000
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 9.884934153917997e-05,
+      "loss": 1.1768,
+      "step": 23000
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 9.86663298624003e-05,
+      "loss": 1.0943,
+      "step": 24000
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 9.847001329696653e-05,
+      "loss": 1.1477,
+      "step": 25000
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 9.826044551386744e-05,
+      "loss": 1.0749,
+      "step": 26000
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 9.803768380684242e-05,
+      "loss": 1.1274,
+      "step": 27000
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 9.780178907671789e-05,
+      "loss": 1.0561,
+      "step": 28000
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 9.755282581475769e-05,
+      "loss": 1.1045,
+      "step": 29000
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 9.729086208503174e-05,
+      "loss": 1.0489,
+      "step": 30000
+    },
+    {
+      "epoch": 14.01,
+      "learning_rate": 9.701596950580806e-05,
+      "loss": 1.0863,
+      "step": 31000
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 9.672822322997305e-05,
+      "loss": 1.0483,
+      "step": 32000
+    },
+    {
+      "epoch": 15.01,
+      "learning_rate": 9.642770192448536e-05,
+      "loss": 1.0622,
+      "step": 33000
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 9.611448774886924e-05,
+      "loss": 1.0711,
+      "step": 34000
+    },
+    {
+      "epoch": 16.01,
+      "learning_rate": 9.578866633275288e-05,
+      "loss": 1.0189,
+      "step": 35000
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 9.545032675245813e-05,
+      "loss": 1.0806,
+      "step": 36000
+    },
+    {
+      "epoch": 17.01,
+      "learning_rate": 9.509956150664796e-05,
+      "loss": 0.9976,
+      "step": 37000
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 9.473646649103818e-05,
+      "loss": 1.0707,
+      "step": 38000
+    },
+    {
+      "epoch": 18.01,
+      "learning_rate": 9.43611409721806e-05,
+      "loss": 0.9852,
+      "step": 39000
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 9.397368756032445e-05,
+      "loss": 1.057,
+      "step": 40000
+    },
+    {
+      "epoch": 19.01,
+      "learning_rate": 9.357421218136386e-05,
+      "loss": 0.988,
+      "step": 41000
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 9.316282404787871e-05,
+      "loss": 1.0394,
+      "step": 42000
+    },
+    {
+      "epoch": 20.01,
+      "learning_rate": 9.273963562927695e-05,
+      "loss": 0.9825,
+      "step": 43000
+    },
+    {
+      "epoch": 21.0,
+      "learning_rate": 9.230476262104677e-05,
+      "loss": 1.0418,
+      "step": 44000
+    },
+    {
+      "epoch": 21.01,
+      "learning_rate": 9.185832391312644e-05,
+      "loss": 0.9664,
+      "step": 45000
+    },
+    {
+      "epoch": 21.01,
+      "learning_rate": 9.140044155740101e-05,
+      "loss": 1.0387,
+      "step": 46000
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 9.093124073433463e-05,
+      "loss": 0.9689,
+      "step": 47000
+    },
+    {
+      "epoch": 22.01,
+      "learning_rate": 9.045084971874738e-05,
+      "loss": 1.0214,
+      "step": 48000
+    },
+    {
+      "epoch": 23.0,
+      "learning_rate": 8.995939984474624e-05,
+      "loss": 0.9581,
+      "step": 49000
+    },
+    {
+      "epoch": 23.01,
+      "learning_rate": 8.945702546981969e-05,
+      "loss": 1.013,
+      "step": 50000
+    },
+    {
+      "epoch": 23.01,
+      "eval_loss": 0.7245611548423767,
+      "eval_runtime": 16.7835,
+      "eval_samples_per_second": 297.912,
+      "eval_steps_per_second": 2.383,
+      "step": 50000
+    }
+  ],
+  "max_steps": 200000,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 8.423667913331835e+17,
+  "trial_name": null,
+  "trial_params": null
+}

last-checkpoint/training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:dcccc81ee4b8b36e817b0bc1898745f6812010dd864ca4597d8c4f4fbfc25a02
+size 3439

pytorch_model.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3ecccd9eb602c427ec07b2e2e082a8f8a5cac7f4aecb7f1cc99d1c0fc82cd41d
+size 442675755

runs/Feb09_18-26-34_t1v-n-07f0074d-w-0/1675967214.712931/events.out.tfevents.1675967214.t1v-n-07f0074d-w-0.3787133.1 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:35f48cb3f60aa00377604a719401b0770de2457b3255cecb47566ae55ac8305e
+size 5458

runs/Feb09_18-26-34_t1v-n-07f0074d-w-0/events.out.tfevents.1675967214.t1v-n-07f0074d-w-0.3787133.0 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c6d8dfc5f914d6f4454db7bbf721ef32c120f844c877042a4ce2bd0e0b73e420
+size 12023

special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,13 @@

+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "name_or_path": "data/plms/legal-lithuanian-roberta-base",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "special_tokens_map_file": "data/plms/legal-lithuanian-roberta-base_32k/special_tokens_map.json",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:dcccc81ee4b8b36e817b0bc1898745f6812010dd864ca4597d8c4f4fbfc25a02
+size 3439