Initial commit

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 50.0,
+    "perplexity": 2.8322171542330707,
+    "test_accuracy": 0.35489852978435493,
+    "test_loss": 1.0410598516464233,
+    "test_runtime": 55.5978,
+    "test_samples_per_second": 4293.445,
+    "test_steps_per_second": 16.781,
+    "train_loss": 1.2392261905236766,
+    "train_runtime": 25350.8633,
+    "train_samples": 1273104,
+    "train_samples_per_second": 2510.968,
+    "train_steps_per_second": 9.808
+}

config.json

@@ -0,0 +1,32 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 1,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 144,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 128,
+  "pad_token": " ",
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.27.1",
+  "use_cache": true,
+  "vocab_size": 93
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "do_sample": true,
+  "eos_token_id": 1,
+  "length_penalty": 0.0,
+  "max_new_tokens": 128,
+  "min_new_tokens": 1,
+  "num_return_sequences": 100,
+  "pad_token_id": 2,
+  "transformers_version": "4.27.1"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75711e4f676270ff40fcfc23ea74221564ba4dafd829420db85f2ef42cdc11dc
+size 12414333

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "^",
+  "eos_token": "_",
+  "pad_token": " ",
+  "unk_token": "§"
+}

test_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 50.0,
+    "perplexity": 2.8322171542330707,
+    "test_accuracy": 0.35489852978435493,
+    "test_loss": 1.0410598516464233,
+    "test_runtime": 55.5978,
+    "test_samples_per_second": 4293.445,
+    "test_steps_per_second": 16.781
+}

tokenizer.json

@@ -0,0 +1,490 @@
+{
+  "version": "1.0",
+  "truncation": null,
+  "padding": null,
+  "added_tokens": [
+    {
+      "id": 0,
+      "content": "^",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 1,
+      "content": "_",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 2,
+      "content": " ",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 3,
+      "content": "§",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 4,
+      "content": "°",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  ],
+  "normalizer": null,
+  "pre_tokenizer": null,
+  "post_processor": {
+    "type": "TemplateProcessing",
+    "single": [
+      {
+        "SpecialToken": {
+          "id": "^",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "SpecialToken": {
+          "id": "_",
+          "type_id": 0
+        }
+      }
+    ],
+    "pair": [
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "B",
+          "type_id": 1
+        }
+      }
+    ],
+    "special_tokens": {
+      "^": {
+        "id": "^",
+        "ids": [
+          0
+        ],
+        "tokens": [
+          "^"
+        ]
+      },
+      "_": {
+        "id": "_",
+        "ids": [
+          1
+        ],
+        "tokens": [
+          "_"
+        ]
+      }
+    }
+  },
+  "decoder": null,
+  "model": {
+    "type": "Unigram",
+    "unk_id": 3,
+    "vocab": [
+      [
+        "^",
+        0.0
+      ],
+      [
+        "_",
+        0.0
+      ],
+      [
+        " ",
+        0.0
+      ],
+      [
+        "§",
+        0.0
+      ],
+      [
+        "°",
+        0.0
+      ],
+      [
+        "C",
+        -2.2339120063759843
+      ],
+      [
+        "c1",
+        -3.108178186508052
+      ],
+      [
+        "cc",
+        -3.2719370147830347
+      ],
+      [
+        "N",
+        -3.2932537094066436
+      ],
+      [
+        "O",
+        -3.294834188186176
+      ],
+      [
+        "n",
+        -3.3053461817854917
+      ],
+      [
+        ")",
+        -3.3475212288619822
+      ],
+      [
+        "CC",
+        -3.4312206506012544
+      ],
+      [
+        "(",
+        -3.5624560302888746
+      ],
+      [
+        "c(",
+        -3.765813889966134
+      ],
+      [
+        "c",
+        -3.8218193366886393
+      ],
+      [
+        "C(=O)",
+        -3.856958418790814
+      ],
+      [
+        "c3",
+        -3.9254062324983607
+      ],
+      [
+        "=",
+        -3.975334503903719
+      ],
+      [
+        "c2",
+        -4.003423620506245
+      ],
+      [
+        "1",
+        -4.112680698145304
+      ],
+      [
+        "(C)",
+        -4.121209869936163
+      ],
+      [
+        "2",
+        -4.1361761826809555
+      ],
+      [
+        "C1",
+        -4.21768981835341
+      ],
+      [
+        "c1ccc(",
+        -4.251735247120498
+      ],
+      [
+        "-",
+        -4.252411803847931
+      ],
+      [
+        "2)",
+        -4.270562896654599
+      ],
+      [
+        "cc(",
+        -4.276471688527964
+      ],
+      [
+        "C(",
+        -4.3086368671374995
+      ],
+      [
+        "CO",
+        -4.337904719405833
+      ],
+      [
+        "C(=O)N",
+        -4.428846224698317
+      ],
+      [
+        "3)",
+        -4.449247292348714
+      ],
+      [
+        "C2",
+        -4.522375726063721
+      ],
+      [
+        "O)",
+        -4.5371556203299654
+      ],
+      [
+        "CCC",
+        -4.589271525409394
+      ],
+      [
+        "3",
+        -4.724794788518322
+      ],
+      [
+        "C(C)",
+        -4.750491328820496
+      ],
+      [
+        "cc1",
+        -4.774712806549024
+      ],
+      [
+        "cn",
+        -4.786244699312933
+      ],
+      [
+        "NC(=O)",
+        -4.789019365908235
+      ],
+      [
+        "CC1",
+        -4.79366206708262
+      ],
+      [
+        "=O)",
+        -4.843480092939821
+      ],
+      [
+        "C)",
+        -4.8437048190443175
+      ],
+      [
+        "S",
+        -4.850239832224922
+      ],
+      [
+        "n1",
+        -4.864510173862389
+      ],
+      [
+        "c1ccccc1",
+        -4.92499844194878
+      ],
+      [
+        "c2ccc(",
+        -5.024291675089154
+      ],
+      [
+        "[nH]",
+        -5.110243493192227
+      ],
+      [
+        "c4",
+        -5.110439906920133
+      ],
+      [
+        "s",
+        -5.124712107023752
+      ],
+      [
+        "N1",
+        -5.13109013422166
+      ],
+      [
+        "o",
+        -5.144341645447907
+      ],
+      [
+        "F)",
+        -5.173986710079323
+      ],
+      [
+        "N(C",
+        -5.185845166641148
+      ],
+      [
+        "S(=O)(=O)",
+        -5.212557352968634
+      ],
+      [
+        "Cl)",
+        -5.220168182623269
+      ],
+      [
+        "c2ccc",
+        -5.2702619038121234
+      ],
+      [
+        "C(O)",
+        -5.276647874822254
+      ],
+      [
+        "2)cc1",
+        -5.314256517516519
+      ],
+      [
+        "O=C(",
+        -5.316263615262265
+      ],
+      [
+        "c3ccccc3",
+        -5.35319737432803
+      ],
+      [
+        "4",
+        -5.365584577986091
+      ],
+      [
+        "c(Cl)c",
+        -5.396247268723645
+      ],
+      [
+        "C=C",
+        -5.435516000963592
+      ],
+      [
+        "5",
+        -5.521017741806011
+      ],
+      [
+        "N2CC",
+        -5.731120719935269
+      ],
+      [
+        "c(F)c",
+        -5.761758311176596
+      ],
+      [
+        "C(F)(F)F)",
+        -5.816389560959275
+      ],
+      [
+        "[",
+        -5.9137028227376565
+      ],
+      [
+        "]",
+        -5.9137028227376565
+      ],
+      [
+        "c(OC)c",
+        -5.948373907493249
+      ],
+      [
+        "c(-c3cc",
+        -6.076591700432848
+      ],
+      [
+        "Br)",
+        -6.173257390983636
+      ],
+      [
+        "#",
+        -6.406719606169013
+      ],
+      [
+        "[N+](=O)[O-])",
+        -6.4134780734073065
+      ],
+      [
+        "+",
+        -6.459711891957882
+      ],
+      [
+        "F",
+        -6.6661526991253
+      ],
+      [
+        "P",
+        -7.117461698495431
+      ],
+      [
+        "6",
+        -7.298200155096458
+      ],
+      [
+        "B",
+        -8.020755755896921
+      ],
+      [
+        "I",
+        -8.076335551264686
+      ],
+      [
+        "7",
+        -9.064406674700315
+      ],
+      [
+        "H",
+        -9.766986825131063
+      ],
+      [
+        "8",
+        -10.67648343715202
+      ],
+      [
+        "9",
+        -11.6208264490498
+      ],
+      [
+        "%",
+        -13.14094059722758
+      ],
+      [
+        "0",
+        -13.512972359677438
+      ],
+      [
+        "p",
+        -13.73258273972798
+      ],
+      [
+        "l",
+        -17.659453778747075
+      ],
+      [
+        "e",
+        -17.659553778747075
+      ],
+      [
+        "i",
+        -17.659653778747074
+      ],
+      [
+        "r",
+        -17.659753778747074
+      ],
+      [
+        "b",
+        -17.659753778747074
+      ]
+    ]
+  }
+}

tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "bos_token": "^",
+  "eos_token": "_",
+  "model_max_length": 128,
+  "pad_token": " ",
+  "padding_side": "right",
+  "special_tokens_map_file": "../../tokenizers/char_unigram_88/special_tokens_map.json",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "truncation_side": "left",
+  "unk_token": "§"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 50.0,
+    "train_loss": 1.2392261905236766,
+    "train_runtime": 25350.8633,
+    "train_samples": 1273104,
+    "train_samples_per_second": 2510.968,
+    "train_steps_per_second": 9.808
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d7f23e857016ba44427fcedf175d39d5d54474fc70e0d9f748a4969ddc5394f
+size 3643