initial commit

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "kravchenko/uk-mt5-small",
+  "architectures": [
+    "MT5ForConditionalGeneration"
+  ],
+  "d_ff": 1024,
+  "d_kv": 64,
+  "d_model": 512,
+  "decoder_start_token_id": 0,
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "mt5",
+  "num_decoder_layers": 8,
+  "num_heads": 6,
+  "num_layers": 8,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
+  "tokenizer_class": "T5Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.19.4",
+  "use_cache": false,
+  "vocab_size": 30000
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00edb6f414e0a81a28dae3c986dbc722b7391fc59e7679f63fc1bc90f300ed7e
+size 1117153

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2afe48ae31e3a7e68172706b28641fcdb246e3779445daca8bdae0b0862fa98e
+size 299192133

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5bcedb333a411722c0a95ed37041e6df857ca2b31077070e9158a44e1e89285
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67d50f8434430bd7dd79cfa80b767cae15899670ea53d7b5f20c3f3495ff8d45
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa8ea53d190098f2632bd1c44b62731a90a9257a96081183c1a361a24f7ea8ff
+size 804017

tokenizer_config.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "sp_model_kwargs": {}, "special_tokens_map_file": "/root/.cache/huggingface/transformers/7d115a6774b803276ce282b932e5802b0b6329001c4bc36e2b9bc26ec9276c0c.294ebaa4cd17bb284635004c92d2c4d522ec488c828dcce0c2471b6f28e3fe82", "name_or_path": "kravchenko/uk-mt5-small", "tokenizer_class": "T5Tokenizer"}

trainer_state.json

@@ -0,0 +1,91 @@
+{
+  "best_metric": 0.31844088435173035,
+  "best_model_checkpoint": "uk-mt5-small-gec/checkpoint-1000",
+  "epoch": 3.802281368821293,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0009239543726235742,
+      "loss": 1.6591,
+      "step": 200
+    },
+    {
+      "epoch": 0.76,
+      "eval_google_bleu": 0.41810232396378044,
+      "eval_loss": 0.3892889618873596,
+      "eval_runtime": 37.9895,
+      "eval_samples_per_second": 49.224,
+      "eval_steps_per_second": 1.553,
+      "step": 200
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.0008479087452471484,
+      "loss": 0.4151,
+      "step": 400
+    },
+    {
+      "epoch": 1.52,
+      "eval_google_bleu": 0.42063669574293594,
+      "eval_loss": 0.3223673403263092,
+      "eval_runtime": 37.3371,
+      "eval_samples_per_second": 50.084,
+      "eval_steps_per_second": 1.58,
+      "step": 400
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0007718631178707225,
+      "loss": 0.3459,
+      "step": 600
+    },
+    {
+      "epoch": 2.28,
+      "eval_google_bleu": 0.421165716556752,
+      "eval_loss": 0.32651257514953613,
+      "eval_runtime": 37.2905,
+      "eval_samples_per_second": 50.147,
+      "eval_steps_per_second": 1.582,
+      "step": 600
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 0.0006958174904942966,
+      "loss": 0.3001,
+      "step": 800
+    },
+    {
+      "epoch": 3.04,
+      "eval_google_bleu": 0.4227517103515067,
+      "eval_loss": 0.3188486397266388,
+      "eval_runtime": 37.7047,
+      "eval_samples_per_second": 49.596,
+      "eval_steps_per_second": 1.565,
+      "step": 800
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.0006197718631178707,
+      "loss": 0.2519,
+      "step": 1000
+    },
+    {
+      "epoch": 3.8,
+      "eval_google_bleu": 0.42282362881781355,
+      "eval_loss": 0.31844088435173035,
+      "eval_runtime": 37.2519,
+      "eval_samples_per_second": 50.199,
+      "eval_steps_per_second": 1.584,
+      "step": 1000
+    }
+  ],
+  "max_steps": 2630,
+  "num_train_epochs": 10,
+  "total_flos": 2073623000297472.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:746c97bafc69da2b2698f396394f095a9105a530638bde3ce5ecd10af8ac6e76
+size 3247