First model version

config.json

@@ -0,0 +1,60 @@
+{
+  "_name_or_path": "t5-base",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "d_ff": 3072,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "relu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 200,
+      "min_length": 30,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4,
+      "prefix": "summarize: "
+    },
+    "translation_en_to_de": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to German: "
+    },
+    "translation_en_to_fr": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to French: "
+    },
+    "translation_en_to_ro": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to Romanian: "
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.28.0",
+  "use_cache": true,
+  "vocab_size": 32128
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.28.0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:919659982bb070881d4c011e10c4466abf14ec5202f0f03877187bea9890858b
+size 891702929

trainer_state.json

@@ -0,0 +1,185 @@
+{
+  "best_metric": 1.2624495029449463,
+  "epoch": 2.0,
+  "global_step": 12460,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.899678972712681e-06,
+      "loss": 1.5273,
+      "step": 500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.799357945425362e-06,
+      "loss": 1.3678,
+      "step": 1000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.699036918138043e-06,
+      "loss": 1.2869,
+      "step": 1500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.598715890850723e-06,
+      "loss": 1.2951,
+      "step": 2000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.498394863563404e-06,
+      "loss": 1.2578,
+      "step": 2500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.398073836276083e-06,
+      "loss": 1.2532,
+      "step": 3000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.297752808988765e-06,
+      "loss": 1.2441,
+      "step": 3500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.197431781701446e-06,
+      "loss": 1.212,
+      "step": 4000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.097110754414125e-06,
+      "loss": 1.2009,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 8.996789727126807e-06,
+      "loss": 1.1731,
+      "step": 5000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 8.896468699839486e-06,
+      "loss": 1.2008,
+      "step": 5500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 8.796147672552167e-06,
+      "loss": 1.1732,
+      "step": 6000
+    },
+    {
+      "epoch": 2.0,
+      "eval_gen_len": 18.808,
+      "eval_loss": 1.2909361124038696,
+      "eval_rouge1": 38.4799,
+      "eval_rouge2": 14.1438,
+      "eval_rougeL": 32.0853,
+      "eval_rougeLsum": 34.22,
+      "eval_runtime": 725.7342,
+      "eval_samples_per_second": 2.067,
+      "eval_steps_per_second": 1.033,
+      "step": 6230
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 8.69582664526485e-06,
+      "loss": 1.1773,
+      "step": 6500
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 8.595505617977528e-06,
+      "loss": 1.1377,
+      "step": 7000
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 8.49518459069021e-06,
+      "loss": 1.1186,
+      "step": 7500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 8.39486356340289e-06,
+      "loss": 1.106,
+      "step": 8000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 8.29454253611557e-06,
+      "loss": 1.1346,
+      "step": 8500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 8.19422150882825e-06,
+      "loss": 1.1376,
+      "step": 9000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 8.093900481540931e-06,
+      "loss": 1.1118,
+      "step": 9500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 7.993579454253612e-06,
+      "loss": 1.1356,
+      "step": 10000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 7.893258426966293e-06,
+      "loss": 1.0916,
+      "step": 10500
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.792937399678973e-06,
+      "loss": 1.125,
+      "step": 11000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.692616372391654e-06,
+      "loss": 1.1061,
+      "step": 11500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.592295345104335e-06,
+      "loss": 1.1101,
+      "step": 12000
+    },
+    {
+      "epoch": 2.0,
+      "eval_gen_len": 18.808,
+      "eval_loss": 1.2624495029449463,
+      "eval_rouge1": 38.462,
+      "eval_rouge2": 14.3899,
+      "eval_rougeL": 32.0676,
+      "eval_rougeLsum": 34.0319,
+      "eval_runtime": 723.1779,
+      "eval_samples_per_second": 2.074,
+      "eval_steps_per_second": 1.037,
+      "step": 12460
+    }
+  ],
+  "max_steps": 49840,
+  "num_train_epochs": 8,
+  "total_flos": 2.96391223296e+16,
+  "trial_name": null,
+  "trial_params": null
+}