adding model finetuned on tar

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 42.81929990539262,
+    "eval_f1": 61.36379559447102,
+    "eval_samples": 10624,
+    "train_loss": 1.8848772167761034,
+    "train_runtime": 2200.5584,
+    "train_samples": 87747,
+    "train_samples_per_second": 119.625,
+    "train_steps_per_second": 7.478
+}

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "dccuchile/albert-base-2-spanish",
+  "architectures": [
+    "AlbertForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0,
+  "bos_token_id": 2,
+  "classifier_dropout_prob": 0.1,
+  "down_scale_factor": 1,
+  "embedding_size": 128,
+  "eos_token_id": 3,
+  "gap_size": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "inner_group_num": 1,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "albert",
+  "net_structure_type": 0,
+  "num_attention_heads": 12,
+  "num_hidden_groups": 1,
+  "num_hidden_layers": 2,
+  "num_memory_blocks": 0,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1",
+  "type_vocab_size": 2,
+  "vocab_size": 31000
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 42.81929990539262,
+    "eval_f1": 61.36379559447102,
+    "eval_samples": 10624
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:917a386d6b9493a0e4a48299718df33a3fe4d1d49fc98a92ec70d64474e06255
+size 44903231

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "<pad>",
+  "sep_token": "[SEP]",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "eos_token": "[SEP]",
+  "keep_accents": true,
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "[MASK]",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 512,
+  "name_or_path": "dccuchile/albert-base-2-spanish",
+  "pad_token": "<pad>",
+  "remove_space": true,
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "AlbertTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.8848772167761034,
+    "train_runtime": 2200.5584,
+    "train_samples": 87747,
+    "train_samples_per_second": 119.625,
+    "train_steps_per_second": 7.478
+}

trainer_state.json

@@ -0,0 +1,541 @@
+{
+  "best_metric": 61.36379559447102,
+  "best_model_checkpoint": "/home/jcanete/ft-data/all_results/tar/albeto_base_2/epochs_3_bs_16_lr_5e-5/checkpoint-10500",
+  "epoch": 3.0,
+  "global_step": 16455,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "eval_exact_match": 21.75023651844844,
+      "eval_f1": 37.57447600514485,
+      "step": 300
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.848374354299605e-05,
+      "loss": 3.3269,
+      "step": 500
+    },
+    {
+      "epoch": 0.11,
+      "eval_exact_match": 24.66414380321665,
+      "eval_f1": 41.23440962664224,
+      "step": 600
+    },
+    {
+      "epoch": 0.16,
+      "eval_exact_match": 32.44087038789026,
+      "eval_f1": 49.56736430711076,
+      "step": 900
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.6967487085992104e-05,
+      "loss": 2.677,
+      "step": 1000
+    },
+    {
+      "epoch": 0.22,
+      "eval_exact_match": 34.16272469252602,
+      "eval_f1": 52.17403937631078,
+      "step": 1200
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.544819203889396e-05,
+      "loss": 2.4733,
+      "step": 1500
+    },
+    {
+      "epoch": 0.27,
+      "eval_exact_match": 35.496688741721854,
+      "eval_f1": 53.19348275463691,
+      "step": 1500
+    },
+    {
+      "epoch": 0.33,
+      "eval_exact_match": 36.40491958372753,
+      "eval_f1": 54.831302693730166,
+      "step": 1800
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.392889699179581e-05,
+      "loss": 2.3891,
+      "step": 2000
+    },
+    {
+      "epoch": 0.38,
+      "eval_exact_match": 36.74550614947966,
+      "eval_f1": 54.7644398701711,
+      "step": 2100
+    },
+    {
+      "epoch": 0.44,
+      "eval_exact_match": 38.42951750236519,
+      "eval_f1": 55.69824697753973,
+      "step": 2400
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.2409601944697666e-05,
+      "loss": 2.2821,
+      "step": 2500
+    },
+    {
+      "epoch": 0.49,
+      "eval_exact_match": 38.41059602649007,
+      "eval_f1": 56.0614093765921,
+      "step": 2700
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.089030689759952e-05,
+      "loss": 2.2559,
+      "step": 3000
+    },
+    {
+      "epoch": 0.55,
+      "eval_exact_match": 39.33774834437086,
+      "eval_f1": 57.385105012672504,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6,
+      "eval_exact_match": 38.656575212866606,
+      "eval_f1": 56.03551336622864,
+      "step": 3300
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.9371011850501374e-05,
+      "loss": 2.2176,
+      "step": 3500
+    },
+    {
+      "epoch": 0.66,
+      "eval_exact_match": 39.517502365184484,
+      "eval_f1": 57.32723067329097,
+      "step": 3600
+    },
+    {
+      "epoch": 0.71,
+      "eval_exact_match": 40.397350993377486,
+      "eval_f1": 58.23372912964764,
+      "step": 3900
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.785171680340323e-05,
+      "loss": 2.1779,
+      "step": 4000
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 40.14191106906339,
+      "eval_f1": 58.213042409331,
+      "step": 4200
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.6332421756305075e-05,
+      "loss": 2.1644,
+      "step": 4500
+    },
+    {
+      "epoch": 0.82,
+      "eval_exact_match": 40.96499526963103,
+      "eval_f1": 58.918006140859504,
+      "step": 4500
+    },
+    {
+      "epoch": 0.88,
+      "eval_exact_match": 41.324503311258276,
+      "eval_f1": 59.21702126158363,
+      "step": 4800
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.481616529930112e-05,
+      "loss": 2.1601,
+      "step": 5000
+    },
+    {
+      "epoch": 0.93,
+      "eval_exact_match": 41.40964995269631,
+      "eval_f1": 59.44853824992533,
+      "step": 5100
+    },
+    {
+      "epoch": 0.98,
+      "eval_exact_match": 41.97729422894986,
+      "eval_f1": 60.06117101015576,
+      "step": 5400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.329687025220298e-05,
+      "loss": 2.0731,
+      "step": 5500
+    },
+    {
+      "epoch": 1.04,
+      "eval_exact_match": 41.68401135288553,
+      "eval_f1": 59.830806199953315,
+      "step": 5700
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.177757520510483e-05,
+      "loss": 1.8064,
+      "step": 6000
+    },
+    {
+      "epoch": 1.09,
+      "eval_exact_match": 42.02459791863765,
+      "eval_f1": 59.84599471093964,
+      "step": 6000
+    },
+    {
+      "epoch": 1.15,
+      "eval_exact_match": 41.67455061494797,
+      "eval_f1": 59.31725434528457,
+      "step": 6300
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.0258280158006685e-05,
+      "loss": 1.7892,
+      "step": 6500
+    },
+    {
+      "epoch": 1.2,
+      "eval_exact_match": 41.57048249763481,
+      "eval_f1": 60.199256429502846,
+      "step": 6600
+    },
+    {
+      "epoch": 1.26,
+      "eval_exact_match": 41.9678334910123,
+      "eval_f1": 59.90500225716347,
+      "step": 6900
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 2.873898511090854e-05,
+      "loss": 1.7917,
+      "step": 7000
+    },
+    {
+      "epoch": 1.31,
+      "eval_exact_match": 42.42194891201514,
+      "eval_f1": 60.72762889522327,
+      "step": 7200
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.722272865390459e-05,
+      "loss": 1.788,
+      "step": 7500
+    },
+    {
+      "epoch": 1.37,
+      "eval_exact_match": 42.36518448438978,
+      "eval_f1": 60.809943991859576,
+      "step": 7500
+    },
+    {
+      "epoch": 1.42,
+      "eval_exact_match": 42.48817407757805,
+      "eval_f1": 60.79429012527435,
+      "step": 7800
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 2.5703433606806444e-05,
+      "loss": 1.8415,
+      "step": 8000
+    },
+    {
+      "epoch": 1.48,
+      "eval_exact_match": 42.639545884579,
+      "eval_f1": 60.460391744838276,
+      "step": 8100
+    },
+    {
+      "epoch": 1.53,
+      "eval_exact_match": 42.620624408703875,
+      "eval_f1": 61.10136266360258,
+      "step": 8400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 2.4184138559708297e-05,
+      "loss": 1.798,
+      "step": 8500
+    },
+    {
+      "epoch": 1.59,
+      "eval_exact_match": 42.35572374645222,
+      "eval_f1": 60.73530844456789,
+      "step": 8700
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 2.266484351261015e-05,
+      "loss": 1.817,
+      "step": 9000
+    },
+    {
+      "epoch": 1.64,
+      "eval_exact_match": 42.92336802270577,
+      "eval_f1": 60.69895519530886,
+      "step": 9000
+    },
+    {
+      "epoch": 1.7,
+      "eval_exact_match": 42.86660359508041,
+      "eval_f1": 60.66334036689526,
+      "step": 9300
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.11485870556062e-05,
+      "loss": 1.7745,
+      "step": 9500
+    },
+    {
+      "epoch": 1.75,
+      "eval_exact_match": 43.074739829706715,
+      "eval_f1": 61.09186469003939,
+      "step": 9600
+    },
+    {
+      "epoch": 1.8,
+      "eval_exact_match": 42.904446546830656,
+      "eval_f1": 61.13917775897811,
+      "step": 9900
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.9629292008508053e-05,
+      "loss": 1.7932,
+      "step": 10000
+    },
+    {
+      "epoch": 1.86,
+      "eval_exact_match": 42.838221381267736,
+      "eval_f1": 61.00012403857813,
+      "step": 10200
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.8109996961409907e-05,
+      "loss": 1.8128,
+      "step": 10500
+    },
+    {
+      "epoch": 1.91,
+      "eval_exact_match": 42.81929990539262,
+      "eval_f1": 61.36379559447102,
+      "step": 10500
+    },
+    {
+      "epoch": 1.97,
+      "eval_exact_match": 42.50709555345317,
+      "eval_f1": 60.829640531869714,
+      "step": 10800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.6593740504405955e-05,
+      "loss": 1.756,
+      "step": 11000
+    },
+    {
+      "epoch": 2.02,
+      "eval_exact_match": 42.72469252601703,
+      "eval_f1": 61.0949824983516,
+      "step": 11100
+    },
+    {
+      "epoch": 2.08,
+      "eval_exact_match": 42.544938505203405,
+      "eval_f1": 60.824356487676354,
+      "step": 11400
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 1.5074445457307808e-05,
+      "loss": 1.4926,
+      "step": 11500
+    },
+    {
+      "epoch": 2.13,
+      "eval_exact_match": 42.18543046357616,
+      "eval_f1": 60.54564764009016,
+      "step": 11700
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.3555150410209664e-05,
+      "loss": 1.4462,
+      "step": 12000
+    },
+    {
+      "epoch": 2.19,
+      "eval_exact_match": 42.535477767265846,
+      "eval_f1": 60.63187441471661,
+      "step": 12000
+    },
+    {
+      "epoch": 2.24,
+      "eval_exact_match": 42.100283822138124,
+      "eval_f1": 60.76346385090341,
+      "step": 12300
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.2035855363111518e-05,
+      "loss": 1.4906,
+      "step": 12500
+    },
+    {
+      "epoch": 2.3,
+      "eval_exact_match": 42.3368022705771,
+      "eval_f1": 60.88032689469819,
+      "step": 12600
+    },
+    {
+      "epoch": 2.35,
+      "eval_exact_match": 42.29895931882687,
+      "eval_f1": 60.75186551387642,
+      "step": 12900
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 1.051656031601337e-05,
+      "loss": 1.4723,
+      "step": 13000
+    },
+    {
+      "epoch": 2.41,
+      "eval_exact_match": 42.29895931882687,
+      "eval_f1": 61.06192361554967,
+      "step": 13200
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 8.997265268915224e-06,
+      "loss": 1.4917,
+      "step": 13500
+    },
+    {
+      "epoch": 2.46,
+      "eval_exact_match": 42.232734153263955,
+      "eval_f1": 60.9421036656735,
+      "step": 13500
+    },
+    {
+      "epoch": 2.52,
+      "eval_exact_match": 42.39356669820246,
+      "eval_f1": 61.102189737578314,
+      "step": 13800
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 7.4779702218170765e-06,
+      "loss": 1.4979,
+      "step": 14000
+    },
+    {
+      "epoch": 2.57,
+      "eval_exact_match": 42.34626300851466,
+      "eval_f1": 61.05567412635194,
+      "step": 14100
+    },
+    {
+      "epoch": 2.63,
+      "eval_exact_match": 42.8003784295175,
+      "eval_f1": 61.240965933718726,
+      "step": 14400
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 5.95867517471893e-06,
+      "loss": 1.4725,
+      "step": 14500
+    },
+    {
+      "epoch": 2.68,
+      "eval_exact_match": 42.3368022705771,
+      "eval_f1": 60.933088700510005,
+      "step": 14700
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.4424187177149806e-06,
+      "loss": 1.4776,
+      "step": 15000
+    },
+    {
+      "epoch": 2.73,
+      "eval_exact_match": 42.55439924314096,
+      "eval_f1": 60.96455686411175,
+      "step": 15000
+    },
+    {
+      "epoch": 2.79,
+      "eval_exact_match": 42.82876064333018,
+      "eval_f1": 61.07475280902827,
+      "step": 15300
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 2.9231236706168336e-06,
+      "loss": 1.4236,
+      "step": 15500
+    },
+    {
+      "epoch": 2.84,
+      "eval_exact_match": 42.57332071901608,
+      "eval_f1": 61.128070584216076,
+      "step": 15600
+    },
+    {
+      "epoch": 2.9,
+      "eval_exact_match": 42.70577105014191,
+      "eval_f1": 61.08493602964073,
+      "step": 15900
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 1.4038286235186874e-06,
+      "loss": 1.4909,
+      "step": 16000
+    },
+    {
+      "epoch": 2.95,
+      "eval_exact_match": 42.82876064333018,
+      "eval_f1": 61.181108132512826,
+      "step": 16200
+    },
+    {
+      "epoch": 3.0,
+      "step": 16455,
+      "total_flos": 3819871314614016.0,
+      "train_loss": 1.8848772167761034,
+      "train_runtime": 2200.5584,
+      "train_samples_per_second": 119.625,
+      "train_steps_per_second": 7.478
+    }
+  ],
+  "max_steps": 16455,
+  "num_train_epochs": 3,
+  "total_flos": 3819871314614016.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d03da4589d9dc517dcf05cb563bd10d8671b543d479dd12b01e2a80baed3e67b
+size 3515