Upload folder using huggingface_hub

README.md

@@ -0,0 +1,89 @@
+---
+tags:
+- generated_from_trainer
+datasets:
+- roneneldan/TinyStories
+metrics:
+- accuracy
+model-index:
+- name: gpt2_m020_tiny-stories_1024
+  results:
+  - task:
+      name: Causal Language Modeling
+      type: text-generation
+    dataset:
+      name: roneneldan/TinyStories
+      type: roneneldan/TinyStories
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6759859395777024
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/scads-nlp/morph-gpt_gpt2_tiny-stories/runs/43jyrhid)
+# gpt2_m020_tiny-stories_1024
+
+This model is a fine-tuned version of [](https://huggingface.co/) on the roneneldan/TinyStories dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.2217
+- Accuracy: 0.6760
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 1.0
+
+### Training results
+
+| Training Loss | Epoch  | Step  | Validation Loss | Accuracy |
+|:-------------:|:------:|:-----:|:---------------:|:--------:|
+| 2.9424        | 0.0525 | 1000  | 2.4920          | 0.4375   |
+| 1.9972        | 0.1050 | 2000  | 1.8161          | 0.5644   |
+| 1.7424        | 0.1575 | 3000  | 1.6275          | 0.5979   |
+| 1.6233        | 0.2100 | 4000  | 1.5269          | 0.6162   |
+| 1.5467        | 0.2625 | 5000  | 1.4625          | 0.6284   |
+| 1.4918        | 0.3150 | 6000  | 1.4187          | 0.6369   |
+| 1.4537        | 0.3675 | 7000  | 1.3823          | 0.6438   |
+| 1.4228        | 0.4200 | 8000  | 1.3529          | 0.6496   |
+| 1.3987        | 0.4725 | 9000  | 1.3283          | 0.6543   |
+| 1.378         | 0.5250 | 10000 | 1.3099          | 0.6578   |
+| 1.3567        | 0.5775 | 11000 | 1.2938          | 0.6611   |
+| 1.3436        | 0.6300 | 12000 | 1.2784          | 0.6643   |
+| 1.328         | 0.6825 | 13000 | 1.2664          | 0.6666   |
+| 1.3157        | 0.7350 | 14000 | 1.2546          | 0.6690   |
+| 1.3052        | 0.7875 | 15000 | 1.2462          | 0.6707   |
+| 1.2966        | 0.8400 | 16000 | 1.2370          | 0.6726   |
+| 1.2943        | 0.8925 | 17000 | 1.2296          | 0.6742   |
+| 1.2837        | 0.9450 | 18000 | 1.2244          | 0.6754   |
+| 1.2825        | 0.9975 | 19000 | 1.2217          | 0.6760   |
+
+
+### Framework versions
+
+- Transformers 4.42.3
+- Pytorch 2.2.2+cu121
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.6759859395777024,
+    "eval_loss": 1.221665620803833,
+    "eval_runtime": 52.5898,
+    "eval_samples": 6128,
+    "eval_samples_per_second": 116.524,
+    "eval_steps_per_second": 3.651,
+    "perplexity": 3.39283420511172,
+    "total_flos": 3.18502967574528e+17,
+    "train_loss": 1.5702329817773064,
+    "train_runtime": 7426.3864,
+    "train_samples": 609477,
+    "train_samples_per_second": 82.069,
+    "train_steps_per_second": 2.565
+}

config.json

@@ -0,0 +1,31 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1024,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 1024,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "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": "4.42.3",
+  "use_cache": true,
+  "vocab_size": 50257
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.6759859395777024,
+    "eval_loss": 1.221665620803833,
+    "eval_runtime": 52.5898,
+    "eval_samples": 6128,
+    "eval_samples_per_second": 116.524,
+    "eval_steps_per_second": 3.651,
+    "perplexity": 3.39283420511172
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1024,
+  "eos_token_id": 1024,
+  "transformers_version": "4.42.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b4a4fef21424803410f39f7f7216be711d2d179a3baa1c00f15f1f4d4af9a66
+size 497774208

special_tokens_map.json

@@ -0,0 +1 @@
+{}

tokenizer.json

@@ -0,0 +1,1919 @@
+{
+  "version": "1.0",
+  "truncation": null,
+  "padding": null,
+  "added_tokens": [
+    {
+      "id": 1024,
+      "content": "<|endoftext|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": true,
+      "special": true
+    }
+  ],
+  "normalizer": null,
+  "pre_tokenizer": {
+    "type": "ByteLevel",
+    "add_prefix_space": false,
+    "trim_offsets": true,
+    "use_regex": true
+  },
+  "post_processor": {
+    "type": "ByteLevel",
+    "add_prefix_space": true,
+    "trim_offsets": false,
+    "use_regex": true
+  },
+  "decoder": {
+    "type": "ByteLevel",
+    "add_prefix_space": true,
+    "trim_offsets": true,
+    "use_regex": true
+  },
+  "model": {
+    "type": "BPE",
+    "dropout": null,
+    "unk_token": null,
+    "continuing_subword_prefix": null,
+    "end_of_word_suffix": null,
+    "fuse_unk": false,
+    "byte_fallback": false,
+    "ignore_merges": false,
+    "vocab": {
+      "!": 0,
+      "\"": 1,
+      "#": 2,
+      "$": 3,
+      "%": 4,
+      "&": 5,
+      "'": 6,
+      "(": 7,
+      ")": 8,
+      "*": 9,
+      "+": 10,
+      ",": 11,
+      "-": 12,
+      ".": 13,
+      "/": 14,
+      "0": 15,
+      "1": 16,
+      "2": 17,
+      "3": 18,
+      "4": 19,
+      "5": 20,
+      "6": 21,
+      "7": 22,
+      "8": 23,
+      "9": 24,
+      ":": 25,
+      ";": 26,
+      "<": 27,
+      "=": 28,
+      ">": 29,
+      "?": 30,
+      "@": 31,
+      "A": 32,
+      "B": 33,
+      "C": 34,
+      "D": 35,
+      "E": 36,
+      "F": 37,
+      "G": 38,
+      "H": 39,
+      "I": 40,
+      "J": 41,
+      "K": 42,
+      "L": 43,
+      "M": 44,
+      "N": 45,
+      "O": 46,
+      "P": 47,
+      "Q": 48,
+      "R": 49,
+      "S": 50,
+      "T": 51,
+      "U": 52,
+      "V": 53,
+      "W": 54,
+      "X": 55,
+      "Y": 56,
+      "Z": 57,
+      "[": 58,
+      "\\": 59,
+      "]": 60,
+      "_": 61,
+      "`": 62,
+      "a": 63,
+      "b": 64,
+      "c": 65,
+      "d": 66,
+      "e": 67,
+      "f": 68,
+      "g": 69,
+      "h": 70,
+      "i": 71,
+      "j": 72,
+      "k": 73,
+      "l": 74,
+      "m": 75,
+      "n": 76,
+      "o": 77,
+      "p": 78,
+      "q": 79,
+      "r": 80,
+      "s": 81,
+      "t": 82,
+      "u": 83,
+      "v": 84,
+      "w": 85,
+      "x": 86,
+      "y": 87,
+      "z": 88,
+      "{": 89,
+      "|": 90,
+      "}": 91,
+      "~": 92,
+      "¡": 93,
+      "¢": 94,
+      "£": 95,
+      "¤": 96,
+      "¥": 97,
+      "¦": 98,
+      "§": 99,
+      "¨": 100,
+      "©": 101,
+      "ª": 102,
+      "«": 103,
+      "¬": 104,
+      "®": 105,
+      "¯": 106,
+      "°": 107,
+      "±": 108,
+      "³": 109,
+      "´": 110,
+      "µ": 111,
+      "¶": 112,
+      "·": 113,
+      "¸": 114,
+      "¹": 115,
+      "º": 116,
+      "»": 117,
+      "¼": 118,
+      "½": 119,
+      "¾": 120,
+      "¿": 121,
+      "Â": 122,
+      "Ã": 123,
+      "Ä": 124,
+      "Å": 125,
+      "É": 126,
+      "Ê": 127,
+      "Ñ": 128,
+      "Ò": 129,
+      "á": 130,
+      "â": 131,
+      "ã": 132,
+      "ä": 133,
+      "å": 134,
+      "æ": 135,
+      "ç": 136,
+      "è": 137,
+      "é": 138,
+      "î": 139,
+      "ï": 140,
+      "ð": 141,
+      "ĉ": 142,
+      "Ċ": 143,
+      "Ġ": 144,
+      "Ģ": 145,
+      "ģ": 146,
+      "Ĥ": 147,
+      "ĥ": 148,
+      "Ħ": 149,
+      "ħ": 150,
+      "Ĩ": 151,
+      "ĩ": 152,
+      "Ī": 153,
+      "ī": 154,
+      "Ĭ": 155,
+      "ĭ": 156,
+      "Į": 157,
+      "į": 158,
+      "İ": 159,
+      "ı": 160,
+      "IJ": 161,
+      "ij": 162,
+      "Ĵ": 163,
+      "ĵ": 164,
+      "Ķ": 165,
+      "ķ": 166,
+      "ĸ": 167,
+      "ĺ": 168,
+      "Ļ": 169,
+      "ļ": 170,
+      "Ľ": 171,
+      "ľ": 172,
+      "Ŀ": 173,
+      "ŀ": 174,
+      "Ł": 175,
+      "ł": 176,
+      "Ń": 177,
+      "he": 178,
+      "Ġt": 179,
+      "Ġa": 180,
+      "Ġs": 181,
+      "nd": 182,
+      "Ġw": 183,
+      "Ġthe": 184,
+      "ed": 185,
+      "Ġb": 186,
+      "Ġto": 187,
+      "ĠT": 188,
+      "Ġand": 189,
+      "Ġh": 190,
+      "Ġf": 191,
+      "in": 192,
+      "Ġwa": 193,
+      "re": 194,
+      "it": 195,
+      "ou": 196,
+      "Ġl": 197,
+      "Ġd": 198,
+      "Ġc": 199,
+      "Ġp": 200,
+      "ay": 201,
+      "Ġm": 202,
+      "er": 203,
+      "Ġwas": 204,
+      "om": 205,
+      "Ġhe": 206,
+      "ĠThe": 207,
+      "is": 208,
+      "im": 209,
+      "ar": 210,
+      "Ġn": 211,
+      "on": 212,
+      "Ġsa": 213,
+      "ĠS": 214,
+      "id": 215,
+      "ll": 216,
+      "Ġha": 217,
+      "Ġg": 218,
+      "at": 219,
+      "ing": 220,
+      "ot": 221,
+      "en": 222,
+      "an": 223,
+      "le": 224,
+      "end": 225,
+      "or": 226,
+      "Ġ\"": 227,
+      "ir": 228,
+      "of": 229,
+      "ĠH": 230,
+      "am": 231,
+      "et": 232,
+      "Ġit": 233,
+      "Ġth": 234,
+      "ig": 235,
+      "il": 236,
+      "ĠHe": 237,
+      "Ġin": 238,
+      "Ġpl": 239,
+      "ow": 240,
+      "ver": 241,
+      "ri": 242,
+      "ĠO": 243,
+      "ut": 244,
+      "ĠThey": 245,
+      "Ġbe": 246,
+      "Ġu": 247,
+      "Ġplay": 248,
+      "Ġsaid": 249,
+      "ith": 250,
+      "pp": 251,
+      "Ġwith": 252,
+      "Ġday": 253,
+      "Ġy": 254,
+      "oo": 255,
+      "ex": 256,
+      "Ġr": 257,
+      "ce": 258,
+      "ĠI": 259,
+      "ck": 260,
+      "Ġher": 261,
+      "ld": 262,
+      "Ġhis": 263,
+      "ĠL": 264,
+      "ke": 265,
+      "Ġst": 266,
+      "Ġbig": 267,
+      "nt": 268,
+      "very": 269,
+      "Ġyou": 270,
+      "st": 271,
+      "ĠB": 272,
+      "ĠTim": 273,
+      "ne": 274,
+      "ve": 275,
+      "ked": 276,
+      "ext": 277,
+      "Ġhapp": 278,
+      "Ġon": 279,
+      "ĠShe": 280,
+      "un": 281,
+      "ĠM": 282,
+      "ily": 283,
+      "text": 284,
+      "all": 285,
+      "riend": 286,
+      "Ġfriend": 287,
+      "Ġthey": 288,
+      "Ġli": 289,
+      "Ġwe": 290,
+      "Ġhad": 291,
+      "Ġnot": 292,
+      "Ġup": 293,
+      "her": 294,
+      "Ġwant": 295,
+      "Ġof": 296,
+      "ad": 297,
+      "Ġ<": 298,
+      "se": 299,
+      "|>": 300,
+      "Ġ<|": 301,
+      "itt": 302,
+      "Ġe": 303,
+      "Ġdo": 304,
+      "Ġhappy": 305,
+      "ĠA": 306,
+      "ent": 307,
+      "Ġvery": 308,
+      "Ġthat": 309,
+      "Ġsaw": 310,
+      "oftext": 311,
+      "'s": 312,
+      "ould": 313,
+      "Ġmom": 314,
+      "Ġfor": 315,
+      "Ġsh": 316,
+      "es": 317,
+      "ittle": 318,
+      "Ġlittle": 319,
+      "Ġshe": 320,
+      "ime": 321,
+      "ch": 322,
+      "Ġk": 323,
+      "endoftext": 324,
+      "Ġnam": 325,
+      ".\"": 326,
+      "Ġtime": 327,
+      "ound": 328,
+      "Ġso": 329,
+      "Ġthere": 330,
+      "Ġnamed": 331,
+      "Ġbo": 332,
+      "Ġwere": 333,
+      "Ġne": 334,
+      "ĠLily": 335,
+      "Ġwanted": 336,
+      "out": 337,
+      "Ġbut": 338,
+      "ĠOne": 339,
+      "nce": 340,
+      "Ġfriends": 341,
+      "!\"": 342,
+      "ht": 343,
+      "ird": 344,
+      "ved": 345,
+      "Ġsm": 346,
+      "Ġan": 347,
+      "al": 348,
+      "Ġbird": 349,
+      "ĠTom": 350,
+      "el": 351,
+      "ake": 352,
+      "ue": 353,
+      "ug": 354,
+      "ome": 355,
+      "ĠIt": 356,
+      "Ġtoo": 357,
+      "ĠĊ": 358,
+      "ide": 359,
+      "Ġhel": 360,
+      "Ġwent": 361,
+      "Ġwh": 362,
+      "Ġhelp": 363,
+      "Ġis": 364,
+      "Ġall": 365,
+      "Ġloo": 366,
+      "Ġlo": 367,
+      "ry": 368,
+      "ter": 369,
+      "Ġupon": 370,
+      "ore": 371,
+      "ill": 372,
+      "ame": 373,
+      "ind": 374,
+      "Ġfun": 375,
+      "Ġtoy": 376,
+      "get": 377,
+      "ra": 378,
+      "Ġas": 379,
+      "Ġdid": 380,
+      "Ġat": 381,
+      "Ġj": 382,
+      "Ġre": 383,
+      "gether": 384,
+      "ur": 385,
+      "Ġo": 386,
+      "ack": 387,
+      "Ġse": 388,
+      "ly": 389,
+      "Ġtogether": 390,
+      "Ġtre": 391,
+      "Ġcat": 392,
+      "ĠOnce": 393,
+      "ood": 394,
+      "Ġcould": 395,
+      "ted": 396,
+      "Ġdog": 397,
+      "Ġcan": 398,
+      "Ġtheir": 399,
+      "ec": 400,
+      "ard": 401,
+      "ark": 402,
+      "my": 403,
+      "Ġgir": 404,
+      "Ġplayed": 405,
+      "Ġhim": 406,
+      "Ġball": 407,
+      "Ġro": 408,
+      "?\"": 409,
+      "Ġgirl": 410,
+      "ax": 411,
+      "way": 412,
+      "Ġgo": 413,
+      "um": 414,
+      "Ġare": 415,
+      "Ġout": 416,
+      "Ġle": 417,
+      "Ġfr": 418,
+      "ain": 419,
+      "hen": 420,
+      "ĠW": 421,
+      "'t": 422,
+      "Ġthem": 423,
+      "Ġsad": 424,
+      "ul": 425,
+      "Ġboy": 426,
+      "Ġtree": 427,
+      "other": 428,
+      "Ġhave": 429,
+      "Ġman": 430,
+      "hed": 431,
+      "Ġcl": 432,
+      "oug": 433,
+      "Ġloved": 434,
+      "Ġlooked": 435,
+      "ic": 436,
+      "Ġfound": 437,
+      "Ġsp": 438,
+      "one": 439,
+      "Ġstar": 440,
+      "Ġback": 441,
+      "Ġsc": 442,
+      "hing": 443,
+      "own": 444,
+      "Ġlike": 445,
+      "are": 446,
+      "ful": 447,
+      "ĠJ": 448,
+      "side": 449,
+      "Ġme": 450,
+      "ĠBut": 451,
+      "ight": 452,
+      "Ġla": 453,
+      "ong": 454,
+      "Ġcar": 455,
+      "Ġpark": 456,
+      "op": 457,
+      "ick": 458,
+      "Ġwould": 459,
+      "elt": 460,
+      "ĠSue": 461,
+      "Ġmake": 462,
+      "ell": 463,
+      "round": 464,
+      "Ġfa": 465,
+      "Ġfelt": 466,
+      "Ġsee": 467,
+      "Ġno": 468,
+      "Ġasked": 469,
+      "ag": 470,
+      "omet": 471,
+      "ĠMax": 472,
+      "Ġnew": 473,
+      "ice": 474,
+      "ĠF": 475,
+      "ouse": 476,
+      "Ġstarted": 477,
+      "Ġcame": 478,
+      "Ġother": 479,
+      "ared": 480,
+      "Ġsay": 481,
+      "Ġal": 482,
+      "ss": 483,
+      "pot": 484,
+      "ob": 485,
+      "ought": 486,
+      "Ġsomet": 487,
+      "Ġag": 488,
+      "Ġgood": 489,
+      "Ġsmall": 490,
+      "Ġbr": 491,
+      "ia": 492,
+      "ĠBen": 493,
+      "ade": 494,
+      "Ġbec": 495,
+      "ĠSam": 496,
+      "ings": 497,
+      "ried": 498,
+      "Ġwor": 499,
+      "Ġfind": 500,
+      "Ġsmil": 501,
+      "Ġex": 502,
+      "Ġwal": 503,
+      "Ġaway": 504,
+      "Ġput": 505,
+      "Ġliked": 506,
+      "ty": 507,
+      "ook": 508,
+      "Ġwhat": 509,
+      "Ġco": 510,
+      "Ġsomething": 511,
+      "Ġfrom": 512,
+      "Ġmade": 513,
+      "Ġthought": 514,
+      "Ġhome": 515,
+      "Ġevery": 516,
+      "You": 517,
+      "Ġplaying": 518,
+      "Ġlaug": 519,
+      "Ġmu": 520,
+      "ach": 521,
+      "ile": 522,
+      "uc": 523,
+      "arn": 524,
+      "ie": 525,
+      "Ġran": 526,
+      "ave": 527,
+      "Ġsome": 528,
+      "Ġagain": 529,
+      "now": 530,
+      "Ġfl": 531,
+      "ny": 532,
+      "Ġhouse": 533,
+      "Ġdown": 534,
+      "king": 535,
+      "ure": 536,
+      "Ġtook": 537,
+      "Ġscared": 538,
+      "Ġpr": 539,
+      "Ġtoys": 540,
+      "Ġlearn": 541,
+      "Ġsmiled": 542,
+      "if": 543,
+      "ened": 544,
+      "Ġwill": 545,
+      "Ġbox": 546,
+      "dd": 547,
+      "ret": 548,
+      "ab": 549,
+      "ep": 550,
+      "Ġmy": 551,
+      "uck": 552,
+      "Ġbl": 553,
+      "Ġthings": 554,
+      "Ġaround": 555,
+      "Ġyour": 556,
+      "oud": 557,
+      "ish": 558,
+      "Ġfe": 559,
+      "Ġlived": 560,
+      "Ġsun": 561,
+      "Ġthen": 562,
+      "ank": 563,
+      "as": 564,
+      "ĠSpot": 565,
+      ",\"": 566,
+      "us": 567,
+      "Ġwhen": 568,
+      "Ġch": 569,
+      "hat": 570,
+      "Ġlot": 571,
+      "Ġsw": 572,
+      "ump": 573,
+      "Ġab": 574,
+      "ĠSo": 575,
+      "ĠD": 576,
+      "pped": 577,
+      "Ġget": 578,
+      "ist": 579,
+      "ucy": 580,
+      "ĠLucy": 581,
+      "ap": 582,
+      "oth": 583,
+      "ust": 584,
+      "Ġtried": 585,
+      "Ġknow": 586,
+      "Ġgot": 587,
+      "Ġdec": 588,
+      "Ġwho": 589,
+      "ĠMia": 590,
+      "new": 591,
+      "Ġmany": 592,
+      "nder": 593,
+      "Ġany": 594,
+      "Ġint": 595,
+      "Ġabout": 596,
+      "ited": 597,
+      "Ġpret": 598,
+      "au": 599,
+      "ause": 600,
+      "ally": 601,
+      "ive": 602,
+      "ĠYou": 603,
+      "Ġdecid": 604,
+      "Ġred": 605,
+      "ace": 606,
+      "fter": 607,
+      "Ġmore": 608,
+      "ous": 609,
+      "ise": 610,
+      "ĠMom": 611,
+      "Ġv": 612,
+      "ĠE": 613,
+      "qu": 614,
+      "Ġcare": 615,
+      "Ġhug": 616,
+      "Ġpo": 617,
+      "Ġshow": 618,
+      "Ġlearned": 619,
+      "Ġwater": 620,
+      "urp": 621,
+      "Ġbest": 622,
+      "Ġlook": 623,
+      "Ġjump": 624,
+      "Ġun": 625,
+      "Ġop": 626,
+      "Ġgre": 627,
+      "Ġexc": 628,
+      "ways": 629,
+      "âĢ": 630,
+      "Ġoutside": 631,
+      "Ġalways": 632,
+      "ant": 633,
+      "fe": 634,
+      "Ġroom": 635,
+      "Ġinto": 636,
+      "Ġho": 637,
+      "Ġeat": 638,
+      "nn": 639,
+      "ite": 640,
+      "Ġpe": 641,
+      "Ġboth": 642,
+      "Ġdad": 643,
+      "Ġke": 644,
+      "ower": 645,
+      "Ġsays": 646,
+      "Ġone": 647,
+      "Ġfeel": 648,
+      "Ġexcited": 649,
+      "Ġnice": 650,
+      "Ġthis": 651,
+      "ĠBob": 652,
+      "Ġfast": 653,
+      "udd": 654,
+      "Ġrun": 655,
+      "urpr": 656,
+      "Ġlong": 657,
+      "Ġsurpr": 658,
+      "nna": 659,
+      "Ġsor": 660,
+      "Ġam": 661,
+      "Ġtold": 662,
+      "Ġsk": 663,
+      "our": 664,
+      "Ġmo": 665,
+      "ĠC": 666,
+      "ara": 667,
+      "Ġinside": 668,
+      "Ġtr": 669,
+      "ull": 670,
+      "ink": 671,
+      "Ġknew": 672,
+      "imal": 673,
+      "Ġanimal": 674,
+      "Ġthan": 675,
+      "Ġpretty": 676,
+      "iny": 677,
+      "og": 678,
+      "Ġrock": 679,
+      "Ġgra": 680,
+      "Ġpick": 681,
+      "Ġtake": 682,
+      "hank": 683,
+      "ĠAnd": 684,
+      "Ġeach": 685,
+      "Ġgave": 686,
+      "Ġbecame": 687,
+      "Ġsl": 688,
+      "Ġmuch": 689,
+      "Ġstr": 690,
+      "Ġhow": 691,
+      "Ġneed": 692,
+      "Ġlaughed": 693,
+      "Ġtow": 694,
+      "ven": 695,
+      "etter": 696,
+      "Ġor": 697,
+      "ĠAs": 698,
+      "Ġdecided": 699,
+      "Ġbecause": 700,
+      "Ġunder": 701,
+      "lew": 702,
+      "ged": 703,
+      "Ġold": 704,
+      "ĠThen": 705,
+      "ge": 706,
+      "ro": 707,
+      "here": 708,
+      "rom": 709,
+      "urt": 710,
+      "Ġcle": 711,
+      "Ġfish": 712,
+      "ase": 713,
+      "and": 714,
+      "Ġflower": 715,
+      "Ġclo": 716,
+      "ast": 717,
+      "Ġbear": 718,
+      "ĠWhen": 719,
+      "ess": 720,
+      "Yes": 721,
+      "Ġlist": 722,
+      "Ġhand": 723,
+      "pl": 724,
+      "urn": 725,
+      "Ġkind": 726,
+      "ĠWe": 727,
+      "Ġtry": 728,
+      "Ġjust": 729,
+      "Ġte": 730,
+      "Ġfood": 731,
+      "Ġnear": 732,
+      "Ġhig": 733,
+      "ine": 734,
+      "Ġwat": 735,
+      "Ġide": 736,
+      "ĠIn": 737,
+      "Ġsky": 738,
+      "Ġfi": 739,
+      "ĠHis": 740,
+      "ving": 741,
+      "pec": 742,
+      "Ġus": 743,
+      "more": 744,
+      "Ġidea": 745,
+      "Ġbetter": 746,
+      "Ġtw": 747,
+      "Ġbug": 748,
+      "Ġheard": 749,
+      "gry": 750,
+      "Ġits": 751,
+      "ate": 752,
+      "Ġen": 753,
+      "able": 754,
+      "Ġlet": 755,
+      "ff": 756,
+      "ber": 757,
+      "Ġcareful": 758,
+      "Ġshare": 759,
+      "Ġsurpris": 760,
+      "lf": 761,
+      "Ġif": 762,
+      "Ġfly": 763,
+      "ĠAnna": 764,
+      "Ġstor": 765,
+      "ĠAmy": 766,
+      "ial": 767,
+      "Ġflew": 768,
+      "ion": 769,
+      "ĠK": 770,
+      "Ġcom": 771,
+      "Ġspec": 772,
+      "Ġdan": 773,
+      "Ġspecial": 774,
+      "ĠSara": 775,
+      "his": 776,
+      "ream": 777,
+      "Ġfo": 778,
+      "Ġnever": 779,
+      "ched": 780,
+      "Ġby": 781,
+      "Ġsorry": 782,
+      "Ġbu": 783,
+      "Ġgr": 784,
+      "Ġwind": 785,
+      "Ġlove": 786,
+      "Ġdon": 787,
+      "Ġcol": 788,
+      "ort": 789,
+      "rm": 790,
+      "Ġtal": 791,
+      "Ġclean": 792,
+      "Ġend": 793,
+      "pected": 794,
+      "expected": 795,
+      "Ġeven": 796,
+      "ople": 797,
+      "Ġmag": 798,
+      "Ġwalked": 799,
+      "Ġhard": 800,
+      "Ġshiny": 801,
+      "Ġover": 802,
+      "Ġcolor": 803,
+      "Ġcake": 804,
+      "ak": 805,
+      "Ġbook": 806,
+      "Ġturn": 807,
+      "Ġfore": 808,
+      "hy": 809,
+      "Ġfam": 810,
+      "Ġbad": 811,
+      "imb": 812,
+      "Ġclimb": 813,
+      "Ġsafe": 814,
+      "Ġafter": 815,
+      "ady": 816,
+      "Ġproud": 817,
+      "bb": 818,
+      "Ġpeople": 819,
+      "Ġhurt": 820,
+      "Ġhappened": 821,
+      "Ġcu": 822,
+      "Ġhigh": 823,
+      "ĠHer": 824,
+      "udden": 825,
+      "Ġdidn": 826,
+      "Ġloud": 827,
+      "Ġcome": 828,
+      "Ġche": 829,
+      "arden": 830,
+      "Ġdoor": 831,
+      "Ġgarden": 832,
+      "Ġanimals": 833,
+      "Ġgl": 834,
+      "Ġopened": 835,
+      "Ġground": 836,
+      "Ġpicked": 837,
+      "ĠP": 838,
+      "Ġim": 839,
+      "Ġgive": 840,
+      "ail": 841,
+      "ild": 842,
+      "ĠFrom": 843,
+      "Ġway": 844,
+      "Ġblue": 845,
+      "'m": 846,
+      "Ġstill": 847,
+      "Ġever": 848,
+      "ĠN": 849,
+      "Ġhugged": 850,
+      "iz": 851,
+      "Ġcall": 852,
+      "Ġfar": 853,
+      "ip": 854,
+      "age": 855,
+      "Ġoff": 856,
+      "Ġmagic": 857,
+      "ough": 858,
+      "Thank": 859,
+      "kay": 860,
+      "ĠEvery": 861,
+      "Ġshould": 862,
+      "Ġanymore": 863,
+      "ane": 864,
+      "Ġpar": 865,
+      "Ġfamily": 866,
+      "ool": 867,
+      "Ġstay": 868,
+      "Ġplace": 869,
+      "Ġkid": 870,
+      "Ġpic": 871,
+      "uddenly": 872,
+      "Ġlots": 873,
+      "Ġqu": 874,
+      "uff": 875,
+      "ock": 876,
+      "ct": 877,
+      "Ġwalk": 878,
+      "Ġnow": 879,
+      "Ġgreat": 880,
+      "em": 881,
+      "ture": 882,
+      "Ġstrong": 883,
+      "be": 884,
+      "dy": 885,
+      "Ġunt": 886,
+      "aut": 887,
+      "Ġforest": 888,
+      "lease": 889,
+      "No": 890,
+      "Ġbra": 891,
+      "Ġfrog": 892,
+      "ers": 893,
+      "Ġbro": 894,
+      "Ġuntil": 895,
+      "Ġbeaut": 896,
+      "opped": 897,
+      "Ġsqu": 898,
+      "Ġstick": 899,
+      "Ġnext": 900,
+      "Ġboat": 901,
+      "Ġapp": 902,
+      "ning": 903,
+      "Ġlisten": 904,
+      "oy": 905,
+      "Ġtra": 906,
+      "hes": 907,
+      "Ġpicture": 908,
+      "Ġkids": 909,
+      "Ġyum": 910,
+      "aking": 911,
+      "Ġbeauti": 912,
+      "Ġbeautiful": 913,
+      "ary": 914,
+      "Ġsurprised": 915,
+      "Ġdra": 916,
+      "Let": 917,
+      "Ġclos": 918,
+      "Ġstory": 919,
+      "ĠR": 920,
+      "Ġrain": 921,
+      "Ġwhile": 922,
+      "Ġimp": 923,
+      "rel": 924,
+      "Ġwo": 925,
+      "Ġtown": 926,
+      "pt": 927,
+      "Ġunexpected": 928,
+      "ress": 929,
+      "Ġsof": 930,
+      "ĠAt": 931,
+      "unny": 932,
+      "Ġbeing": 933,
+      "Ġhat": 934,
+      "ree": 935,
+      "Ġmor": 936,
+      "Ġthanked": 937,
+      "Ġad": 938,
+      "Ġcalled": 939,
+      "dded": 940,
+      "Ġrem": 941,
+      "Ġmet": 942,
+      "ies": 943,
+      "Ġcry": 944,
+      "Ġgame": 945,
+      "by": 946,
+      "Ġopen": 947,
+      "ger": 948,
+      "Ġangry": 949,
+      "Ġtruck": 950,
+      "itty": 951,
+      "Ġsoft": 952,
+      "Ġkeep": 953,
+      "Ġjumped": 954,
+      "vent": 955,
+      "les": 956,
+      "Ġlost": 957,
+      "Ġbed": 958,
+      "self": 959,
+      "uffy": 960,
+      "Ġdoll": 961,
+      "Ġwarm": 962,
+      "oon": 963,
+      "illy": 964,
+      "Ġbrave": 965,
+      "Ġsmile": 966,
+      "ĠJo": 967,
+      "Ġmouse": 968,
+      "Ġate": 969,
+      "Ġde": 970,
+      "It": 971,
+      "fore": 972,
+      "so": 973,
+      "Ġwatch": 974,
+      "iss": 975,
+      "Ġleave": 976,
+      "az": 977,
+      "Ġalso": 978,
+      "Ġfell": 979,
+      "Ġsn": 980,
+      "ĠSuddenly": 981,
+      "Ġgreen": 982,
+      "Ġtwo": 983,
+      "Ġface": 984,
+      "ĠG": 985,
+      "ma": 986,
+      "gan": 987,
+      "bit": 988,
+      "ble": 989,
+      "ĠâĢ": 990,
+      "ĠFl": 991,
+      "abbit": 992,
+      "Ġnoise": 993,
+      "owl": 994,
+      "Ġexpl": 995,
+      "Ġwhere": 996,
+      "irst": 997,
+      "Ġsoon": 998,
+      "Ġrabbit": 999,
+      "Ġslide": 1000,
+      "Ġhole": 1001,
+      "Ġhear": 1002,
+      "Ġpus": 1003,
+      "irrel": 1004,
+      "ĠSally": 1005,
+      "ear": 1006,
+      "ĠAfter": 1007,
+      "Ġbefore": 1008,
+      "Ġsat": 1009,
+      "air": 1010,
+      "Ġcook": 1011,
+      "Ġhappily": 1012,
+      "ĠFluffy": 1013,
+      "Ġsquirrel": 1014,
+      "Ġuse": 1015,
+      "Ġkept": 1016,
+      "ĠâĢľ": 1017,
+      "Hi": 1018,
+      "ĠKitty": 1019,
+      "âĢĿ": 1020,
+      "ired": 1021,
+      "Ġhelped": 1022,
+      "Ġshowed": 1023
+    },
+    "merges": [
+      "h e",
+      "Ġ t",
+      "Ġ a",
+      "Ġ s",
+      "n d",
+      "Ġ w",
+      "Ġt he",
+      "e d",
+      "Ġ b",
+      "Ġt o",
+      "Ġ T",
+      "Ġa nd",
+      "Ġ h",
+      "Ġ f",
+      "i n",
+      "Ġw a",
+      "r e",
+      "i t",
+      "o u",
+      "Ġ l",
+      "Ġ d",
+      "Ġ c",
+      "Ġ p",
+      "a y",
+      "Ġ m",
+      "e r",
+      "Ġwa s",
+      "o m",
+      "Ġ he",
+      "ĠT he",
+      "i s",
+      "i m",
+      "a r",
+      "Ġ n",
+      "o n",
+      "Ġs a",
+      "Ġ S",
+      "i d",
+      "l l",
+      "Ġh a",
+      "Ġ g",
+      "a t",
+      "in g",
+      "o t",
+      "e n",
+      "a n",
+      "l e",
+      "e nd",
+      "o r",
+      "Ġ \"",
+      "i r",
+      "o f",
+      "Ġ H",
+      "a m",
+      "e t",
+      "Ġ it",
+      "Ġt h",
+      "i g",
+      "i l",
+      "ĠH e",
+      "Ġ in",
+      "Ġp l",
+      "o w",
+      "v er",
+      "r i",
+      "Ġ O",
+      "u t",
+      "ĠThe y",
+      "Ġb e",
+      "Ġ u",
+      "Ġpl ay",
+      "Ġsa id",
+      "it h",
+      "p p",
+      "Ġw ith",
+      "Ġd ay",
+      "Ġ y",
+      "o o",
+      "e x",
+      "Ġ r",
+      "c e",
+      "Ġ I",
+      "c k",
+      "Ġhe r",
+      "l d",
+      "Ġh is",
+      "Ġ L",
+      "k e",
+      "Ġs t",
+      "Ġb ig",
+      "n t",
+      "ver y",
+      "Ġy ou",
+      "s t",
+      "Ġ B",
+      "ĠT im",
+      "n e",
+      "v e",
+      "k ed",
+      "ex t",
+      "Ġha pp",
+      "Ġ on",
+      "ĠS he",
+      "u n",
+      "Ġ M",
+      "il y",
+      "t ext",
+      "a ll",
+      "ri end",
+      "Ġf riend",
+      "Ġthe y",
+      "Ġl i",
+      "Ġw e",
+      "Ġha d",
+      "Ġn ot",
+      "Ġu p",
+      "he r",
+      "Ġwa nt",
+      "Ġ of",
+      "a d",
+      "Ġ <",
+      "s e",
+      "| >",
+      "Ġ< |",
+      "it t",
+      "Ġ e",
+      "Ġd o",
+      "Ġhapp y",
+      "Ġ A",
+      "en t",
+      "Ġ very",
+      "Ġth at",
+      "Ġsa w",
+      "of text",
+      "' s",
+      "ou ld",
+      "Ġm om",
+      "Ġf or",
+      "Ġs h",
+      "e s",
+      "itt le",
+      "Ġl ittle",
+      "Ġs he",
+      "im e",
+      "c h",
+      "Ġ k",
+      "end oftext",
+      "Ġn am",
+      ". \"",
+      "Ġt ime",
+      "ou nd",
+      "Ġs o",
+      "Ġthe re",
+      "Ġnam ed",
+      "Ġb o",
+      "Ġwe re",
+      "Ġn e",
+      "ĠL ily",
+      "Ġwant ed",
+      "ou t",
+      "Ġb ut",
+      "ĠO ne",
+      "n ce",
+      "Ġfriend s",
+      "! \"",
+      "h t",
+      "ir d",
+      "v ed",
+      "Ġs m",
+      "Ġa n",
+      "a l",
+      "Ġb ird",
+      "ĠT om",
+      "e l",
+      "a ke",
+      "u e",
+      "u g",
+      "om e",
+      "ĠI t",
+      "Ġto o",
+      "Ġ Ċ",
+      "id e",
+      "Ġhe l",
+      "Ġw ent",
+      "Ġw h",
+      "Ġhel p",
+      "Ġ is",
+      "Ġa ll",
+      "Ġl oo",
+      "Ġl o",
+      "r y",
+      "t er",
+      "Ġup on",
+      "o re",
+      "i ll",
+      "am e",
+      "i nd",
+      "Ġf un",
+      "Ġto y",
+      "g et",
+      "r a",
+      "Ġa s",
+      "Ġd id",
+      "Ġa t",
+      "Ġ j",
+      "Ġ re",
+      "get her",
+      "u r",
+      "Ġ o",
+      "a ck",
+      "Ġs e",
+      "l y",
+      "Ġto gether",
+      "Ġt re",
+      "Ġc at",
+      "ĠO nce",
+      "oo d",
+      "Ġc ould",
+      "t ed",
+      "Ġdo g",
+      "Ġc an",
+      "Ġthe ir",
+      "e c",
+      "ar d",
+      "ar k",
+      "m y",
+      "Ġg ir",
+      "Ġplay ed",
+      "Ġh im",
+      "Ġb all",
+      "Ġr o",
+      "? \"",
+      "Ġgir l",
+      "a x",
+      "w ay",
+      "Ġg o",
+      "u m",
+      "Ġa re",
+      "Ġ out",
+      "Ġl e",
+      "Ġf r",
+      "a in",
+      "he n",
+      "Ġ W",
+      "' t",
+      "Ġthe m",
+      "Ġsa d",
+      "u l",
+      "Ġbo y",
+      "Ġtre e",
+      "ot her",
+      "Ġha ve",
+      "Ġm an",
+      "he d",
+      "Ġc l",
+      "ou g",
+      "Ġlo ved",
+      "Ġloo ked",
+      "i c",
+      "Ġf ound",
+      "Ġs p",
+      "on e",
+      "Ġst ar",
+      "Ġb ack",
+      "Ġs c",
+      "h ing",
+      "ow n",
+      "Ġli ke",
+      "a re",
+      "f ul",
+      "Ġ J",
+      "s ide",
+      "Ġm e",
+      "ĠB ut",
+      "ig ht",
+      "Ġl a",
+      "on g",
+      "Ġc ar",
+      "Ġp ark",
+      "o p",
+      "i ck",
+      "Ġw ould",
+      "el t",
+      "ĠS ue",
+      "Ġm ake",
+      "e ll",
+      "r ound",
+      "Ġf a",
+      "Ġf elt",
+      "Ġse e",
+      "Ġn o",
+      "Ġas ked",
+      "a g",
+      "om et",
+      "ĠM ax",
+      "Ġne w",
+      "i ce",
+      "Ġ F",
+      "ou se",
+      "Ġstar ted",
+      "Ġc ame",
+      "Ġ other",
+      "ar ed",
+      "Ġs ay",
+      "Ġa l",
+      "s s",
+      "p ot",
+      "o b",
+      "oug ht",
+      "Ġs omet",
+      "Ġa g",
+      "Ġg ood",
+      "Ġsm all",
+      "Ġb r",
+      "i a",
+      "ĠB en",
+      "ad e",
+      "Ġbe c",
+      "ĠS am",
+      "ing s",
+      "ri ed",
+      "Ġw or",
+      "Ġf ind",
+      "Ġsm il",
+      "Ġ ex",
+      "Ġwa l",
+      "Ġa way",
+      "Ġp ut",
+      "Ġli ked",
+      "t y",
+      "oo k",
+      "Ġwh at",
+      "Ġc o",
+      "Ġsomet hing",
+      "Ġfr om",
+      "Ġm ade",
+      "Ġth ought",
+      "Ġh ome",
+      "Ġe very",
+      "Y ou",
+      "Ġplay ing",
+      "Ġla ug",
+      "Ġm u",
+      "a ch",
+      "i le",
+      "u c",
+      "ar n",
+      "i e",
+      "Ġr an",
+      "a ve",
+      "Ġs ome",
+      "Ġag ain",
+      "n ow",
+      "Ġf l",
+      "n y",
+      "Ġh ouse",
+      "Ġd own",
+      "k ing",
+      "u re",
+      "Ġtoo k",
+      "Ġsc ared",
+      "Ġp r",
+      "Ġtoy s",
+      "Ġle arn",
+      "Ġsmil ed",
+      "i f",
+      "en ed",
+      "Ġw ill",
+      "Ġbo x",
+      "d d",
+      "re t",
+      "a b",
+      "e p",
+      "Ġm y",
+      "u ck",
+      "Ġb l",
+      "Ġth ings",
+      "Ġa round",
+      "Ġyou r",
+      "ou d",
+      "is h",
+      "Ġf e",
+      "Ġli ved",
+      "Ġs un",
+      "Ġthe n",
+      "an k",
+      "a s",
+      "ĠS pot",
+      ", \"",
+      "u s",
+      "Ġw hen",
+      "Ġc h",
+      "h at",
+      "Ġl ot",
+      "Ġs w",
+      "um p",
+      "Ġa b",
+      "ĠS o",
+      "Ġ D",
+      "pp ed",
+      "Ġg et",
+      "is t",
+      "uc y",
+      "ĠL ucy",
+      "a p",
+      "ot h",
+      "u st",
+      "Ġt ried",
+      "Ġk now",
+      "Ġg ot",
+      "Ġd ec",
+      "Ġwh o",
+      "ĠM ia",
+      "ne w",
+      "Ġman y",
+      "nd er",
+      "Ġan y",
+      "Ġin t",
+      "Ġab out",
+      "it ed",
+      "Ġp ret",
+      "a u",
+      "au se",
+      "all y",
+      "i ve",
+      "Ġ You",
+      "Ġdec id",
+      "Ġr ed",
+      "a ce",
+      "f ter",
+      "Ġm ore",
+      "ou s",
+      "is e",
+      "ĠM om",
+      "Ġ v",
+      "Ġ E",
+      "q u",
+      "Ġc are",
+      "Ġh ug",
+      "Ġp o",
+      "Ġsh ow",
+      "Ġlearn ed",
+      "Ġwa ter",
+      "ur p",
+      "Ġbe st",
+      "Ġloo k",
+      "Ġj ump",
+      "Ġu n",
+      "Ġo p",
+      "Ġg re",
+      "Ġex c",
+      "way s",
+      "â Ģ",
+      "Ġout side",
+      "Ġal ways",
+      "an t",
+      "f e",
+      "Ġro om",
+      "Ġint o",
+      "Ġh o",
+      "Ġe at",
+      "n n",
+      "it e",
+      "Ġp e",
+      "Ġb oth",
+      "Ġd ad",
+      "Ġ ke",
+      "ow er",
+      "Ġsay s",
+      "Ġon e",
+      "Ġfe el",
+      "Ġexc ited",
+      "Ġn ice",
+      "Ġth is",
+      "ĠB ob",
+      "Ġfa st",
+      "u dd",
+      "Ġr un",
+      "urp r",
+      "Ġl ong",
+      "Ġs urpr",
+      "nn a",
+      "Ġs or",
+      "Ġa m",
+      "Ġto ld",
+      "Ġs k",
+      "ou r",
+      "Ġm o",
+      "Ġ C",
+      "ar a",
+      "Ġin side",
+      "Ġt r",
+      "u ll",
+      "in k",
+      "Ġk new",
+      "im al",
+      "Ġan imal",
+      "Ġth an",
+      "Ġpret ty",
+      "in y",
+      "o g",
+      "Ġro ck",
+      "Ġg ra",
+      "Ġp ick",
+      "Ġt ake",
+      "h ank",
+      "ĠA nd",
+      "Ġe ach",
+      "Ġg ave",
+      "Ġbec ame",
+      "Ġs l",
+      "Ġmu ch",
+      "Ġst r",
+      "Ġh ow",
+      "Ġne ed",
+      "Ġlaug hed",
+      "Ġto w",
+      "v en",
+      "et ter",
+      "Ġ or",
+      "ĠA s",
+      "Ġdecid ed",
+      "Ġbec ause",
+      "Ġu nder",
+      "le w",
+      "g ed",
+      "Ġo ld",
+      "ĠThe n",
+      "g e",
+      "r o",
+      "he re",
+      "r om",
+      "ur t",
+      "Ġc le",
+      "Ġf ish",
+      "a se",
+      "a nd",
+      "Ġfl ower",
+      "Ġcl o",
+      "a st",
+      "Ġbe ar",
+      "ĠW hen",
+      "es s",
+      "Y es",
+      "Ġl ist",
+      "Ġha nd",
+      "p l",
+      "ur n",
+      "Ġk ind",
+      "ĠW e",
+      "Ġt ry",
+      "Ġj ust",
+      "Ġt e",
+      "Ġf ood",
+      "Ġne ar",
+      "Ġh ig",
+      "in e",
+      "Ġwa t",
+      "Ġ ide",
+      "ĠI n",
+      "Ġsk y",
+      "Ġf i",
+      "ĠH is",
+      "v ing",
+      "p ec",
+      "Ġu s",
+      "m ore",
+      "Ġide a",
+      "Ġb etter",
+      "Ġt w",
+      "Ġb ug",
+      "Ġhe ard",
+      "g ry",
+      "Ġit s",
+      "at e",
+      "Ġ en",
+      "ab le",
+      "Ġl et",
+      "f f",
+      "b er",
+      "Ġcare ful",
+      "Ġsh are",
+      "Ġsurpr is",
+      "l f",
+      "Ġ if",
+      "Ġf ly",
+      "ĠA nna",
+      "Ġst or",
+      "ĠA my",
+      "i al",
+      "Ġf lew",
+      "i on",
+      "Ġ K",
+      "Ġc om",
+      "Ġsp ec",
+      "Ġd an",
+      "Ġspec ial",
+      "ĠS ara",
+      "h is",
+      "re am",
+      "Ġf o",
+      "Ġne ver",
+      "c hed",
+      "Ġb y",
+      "Ġsor ry",
+      "Ġb u",
+      "Ġg r",
+      "Ġw ind",
+      "Ġlo ve",
+      "Ġd on",
+      "Ġco l",
+      "or t",
+      "r m",
+      "Ġt al",
+      "Ġcle an",
+      "Ġ end",
+      "pec ted",
+      "ex pected",
+      "Ġe ven",
+      "op le",
+      "Ġm ag",
+      "Ġwal ked",
+      "Ġh ard",
+      "Ġsh iny",
+      "Ġo ver",
+      "Ġcol or",
+      "Ġc ake",
+      "a k",
+      "Ġb ook",
+      "Ġt urn",
+      "Ġf ore",
+      "h y",
+      "Ġf am",
+      "Ġb ad",
+      "im b",
+      "Ġcl imb",
+      "Ġsa fe",
+      "Ġa fter",
+      "ad y",
+      "Ġpr oud",
+      "b b",
+      "Ġpe ople",
+      "Ġh urt",
+      "Ġhapp ened",
+      "Ġc u",
+      "Ġhig h",
+      "ĠH er",
+      "udd en",
+      "Ġdid n",
+      "Ġl oud",
+      "Ġc ome",
+      "Ġc he",
+      "ard en",
+      "Ġdo or",
+      "Ġg arden",
+      "Ġanimal s",
+      "Ġg l",
+      "Ġop ened",
+      "Ġg round",
+      "Ġpick ed",
+      "Ġ P",
+      "Ġ im",
+      "Ġg ive",
+      "a il",
+      "il d",
+      "ĠF rom",
+      "Ġwa y",
+      "Ġbl ue",
+      "' m",
+      "Ġst ill",
+      "Ġe ver",
+      "Ġ N",
+      "Ġhug ged",
+      "i z",
+      "Ġc all",
+      "Ġf ar",
+      "i p",
+      "ag e",
+      "Ġof f",
+      "Ġmag ic",
+      "oug h",
+      "T hank",
+      "k ay",
+      "ĠE very",
+      "Ġsh ould",
+      "Ġany more",
+      "an e",
+      "Ġp ar",
+      "Ġfam ily",
+      "oo l",
+      "Ġst ay",
+      "Ġpl ace",
+      "Ġk id",
+      "Ġp ic",
+      "udden ly",
+      "Ġlot s",
+      "Ġ qu",
+      "u ff",
+      "o ck",
+      "c t",
+      "Ġwal k",
+      "Ġn ow",
+      "Ġgre at",
+      "e m",
+      "t ure",
+      "Ġstr ong",
+      "b e",
+      "d y",
+      "Ġu nt",
+      "a ut",
+      "Ġfore st",
+      "le ase",
+      "N o",
+      "Ġb ra",
+      "Ġfr og",
+      "er s",
+      "Ġbr o",
+      "Ġunt il",
+      "Ġbe aut",
+      "o pped",
+      "Ġs qu",
+      "Ġst ick",
+      "Ġn ext",
+      "Ġbo at",
+      "Ġa pp",
+      "n ing",
+      "Ġlist en",
+      "o y",
+      "Ġt ra",
+      "he s",
+      "Ġpic ture",
+      "Ġkid s",
+      "Ġy um",
+      "a king",
+      "Ġbeaut i",
+      "Ġbeauti ful",
+      "ar y",
+      "Ġsurpris ed",
+      "Ġd ra",
+      "L et",
+      "Ġclo s",
+      "Ġstor y",
+      "Ġ R",
+      "Ġr ain",
+      "Ġwh ile",
+      "Ġim p",
+      "re l",
+      "Ġw o",
+      "Ġtow n",
+      "p t",
+      "Ġun expected",
+      "re ss",
+      "Ġs of",
+      "ĠA t",
+      "un ny",
+      "Ġbe ing",
+      "Ġha t",
+      "re e",
+      "Ġm or",
+      "Ġthan ked",
+      "Ġa d",
+      "Ġcall ed",
+      "dd ed",
+      "Ġre m",
+      "Ġm et",
+      "i es",
+      "Ġc ry",
+      "Ġg ame",
+      "b y",
+      "Ġop en",
+      "g er",
+      "Ġan gry",
+      "Ġtr uck",
+      "itt y",
+      "Ġsof t",
+      "Ġke ep",
+      "Ġjump ed",
+      "v ent",
+      "le s",
+      "Ġlo st",
+      "Ġb ed",
+      "se lf",
+      "uff y",
+      "Ġdo ll",
+      "Ġwa rm",
+      "o on",
+      "ill y",
+      "Ġbra ve",
+      "Ġsm ile",
+      "ĠJ o",
+      "Ġm ouse",
+      "Ġat e",
+      "Ġd e",
+      "I t",
+      "f ore",
+      "s o",
+      "Ġwat ch",
+      "is s",
+      "Ġle ave",
+      "a z",
+      "Ġal so",
+      "Ġf ell",
+      "Ġs n",
+      "ĠS uddenly",
+      "Ġgre en",
+      "Ġtw o",
+      "Ġfa ce",
+      "Ġ G",
+      "m a",
+      "g an",
+      "b it",
+      "b le",
+      "Ġ âĢ",
+      "ĠF l",
+      "ab bit",
+      "Ġno ise",
+      "ow l",
+      "Ġex pl",
+      "Ġw here",
+      "ir st",
+      "Ġso on",
+      "Ġr abbit",
+      "Ġsl ide",
+      "Ġho le",
+      "Ġhe ar",
+      "Ġp us",
+      "ir rel",
+      "ĠS ally",
+      "e ar",
+      "ĠA fter",
+      "Ġbe fore",
+      "Ġsa t",
+      "a ir",
+      "Ġc ook",
+      "Ġhapp ily",
+      "ĠFl uffy",
+      "Ġsqu irrel",
+      "Ġu se",
+      "Ġke pt",
+      "ĠâĢ ľ",
+      "H i",
+      "ĠK itty",
+      "âĢ Ŀ",
+      "ir ed",
+      "Ġhelp ed",
+      "Ġshow ed"
+    ]
+  }
+}

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "added_tokens_decoder": {
+    "1024": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 3.18502967574528e+17,
+    "train_loss": 1.5702329817773064,
+    "train_runtime": 7426.3864,
+    "train_samples": 609477,
+    "train_samples_per_second": 82.069,
+    "train_steps_per_second": 2.565
+}

trainer_state.json

@@ -0,0 +1,479 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 1000,
+  "global_step": 19047,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.026250853152727464,
+      "grad_norm": 1.8290232419967651,
+      "learning_rate": 4.868745734236363e-05,
+      "loss": 4.3877,
+      "step": 500
+    },
+    {
+      "epoch": 0.05250170630545493,
+      "grad_norm": 1.7457512617111206,
+      "learning_rate": 4.7374914684727256e-05,
+      "loss": 2.9424,
+      "step": 1000
+    },
+    {
+      "epoch": 0.05250170630545493,
+      "eval_accuracy": 0.4374649063701957,
+      "eval_loss": 2.49202299118042,
+      "eval_runtime": 52.9352,
+      "eval_samples_per_second": 115.764,
+      "eval_steps_per_second": 3.627,
+      "step": 1000
+    },
+    {
+      "epoch": 0.0787525594581824,
+      "grad_norm": 2.0498085021972656,
+      "learning_rate": 4.606237202709088e-05,
+      "loss": 2.3234,
+      "step": 1500
+    },
+    {
+      "epoch": 0.10500341261090985,
+      "grad_norm": 2.0251238346099854,
+      "learning_rate": 4.474982936945451e-05,
+      "loss": 1.9972,
+      "step": 2000
+    },
+    {
+      "epoch": 0.10500341261090985,
+      "eval_accuracy": 0.5644156336378184,
+      "eval_loss": 1.816135287284851,
+      "eval_runtime": 52.6607,
+      "eval_samples_per_second": 116.368,
+      "eval_steps_per_second": 3.646,
+      "step": 2000
+    },
+    {
+      "epoch": 0.13125426576363733,
+      "grad_norm": 2.362931966781616,
+      "learning_rate": 4.343728671181814e-05,
+      "loss": 1.8434,
+      "step": 2500
+    },
+    {
+      "epoch": 0.1575051189163648,
+      "grad_norm": 1.366226315498352,
+      "learning_rate": 4.2124744054181764e-05,
+      "loss": 1.7424,
+      "step": 3000
+    },
+    {
+      "epoch": 0.1575051189163648,
+      "eval_accuracy": 0.5978558749288556,
+      "eval_loss": 1.6274888515472412,
+      "eval_runtime": 52.8918,
+      "eval_samples_per_second": 115.859,
+      "eval_steps_per_second": 3.63,
+      "step": 3000
+    },
+    {
+      "epoch": 0.18375597206909225,
+      "grad_norm": 1.248099446296692,
+      "learning_rate": 4.081220139654539e-05,
+      "loss": 1.6762,
+      "step": 3500
+    },
+    {
+      "epoch": 0.2100068252218197,
+      "grad_norm": 1.2411609888076782,
+      "learning_rate": 3.949965873890902e-05,
+      "loss": 1.6233,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2100068252218197,
+      "eval_accuracy": 0.6161887871386313,
+      "eval_loss": 1.5268635749816895,
+      "eval_runtime": 52.4436,
+      "eval_samples_per_second": 116.849,
+      "eval_steps_per_second": 3.661,
+      "step": 4000
+    },
+    {
+      "epoch": 0.23625767837454717,
+      "grad_norm": 1.3604716062545776,
+      "learning_rate": 3.8187116081272645e-05,
+      "loss": 1.5804,
+      "step": 4500
+    },
+    {
+      "epoch": 0.26250853152727466,
+      "grad_norm": 1.4850926399230957,
+      "learning_rate": 3.687457342363627e-05,
+      "loss": 1.5467,
+      "step": 5000
+    },
+    {
+      "epoch": 0.26250853152727466,
+      "eval_accuracy": 0.6283651600652359,
+      "eval_loss": 1.4624590873718262,
+      "eval_runtime": 52.7312,
+      "eval_samples_per_second": 116.212,
+      "eval_steps_per_second": 3.641,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2887593846800021,
+      "grad_norm": 1.1516271829605103,
+      "learning_rate": 3.55620307659999e-05,
+      "loss": 1.518,
+      "step": 5500
+    },
+    {
+      "epoch": 0.3150102378327296,
+      "grad_norm": 1.2089773416519165,
+      "learning_rate": 3.4249488108363525e-05,
+      "loss": 1.4918,
+      "step": 6000
+    },
+    {
+      "epoch": 0.3150102378327296,
+      "eval_accuracy": 0.6369396504419245,
+      "eval_loss": 1.4186700582504272,
+      "eval_runtime": 52.6824,
+      "eval_samples_per_second": 116.32,
+      "eval_steps_per_second": 3.644,
+      "step": 6000
+    },
+    {
+      "epoch": 0.34126109098545704,
+      "grad_norm": 1.18433678150177,
+      "learning_rate": 3.293694545072715e-05,
+      "loss": 1.4721,
+      "step": 6500
+    },
+    {
+      "epoch": 0.3675119441381845,
+      "grad_norm": 1.1634047031402588,
+      "learning_rate": 3.162440279309078e-05,
+      "loss": 1.4537,
+      "step": 7000
+    },
+    {
+      "epoch": 0.3675119441381845,
+      "eval_accuracy": 0.6438454387214179,
+      "eval_loss": 1.382287859916687,
+      "eval_runtime": 53.055,
+      "eval_samples_per_second": 115.503,
+      "eval_steps_per_second": 3.619,
+      "step": 7000
+    },
+    {
+      "epoch": 0.39376279729091196,
+      "grad_norm": 1.172624111175537,
+      "learning_rate": 3.0311860135454406e-05,
+      "loss": 1.4409,
+      "step": 7500
+    },
+    {
+      "epoch": 0.4200136504436394,
+      "grad_norm": 1.242315649986267,
+      "learning_rate": 2.8999317477818033e-05,
+      "loss": 1.4228,
+      "step": 8000
+    },
+    {
+      "epoch": 0.4200136504436394,
+      "eval_accuracy": 0.649608131768285,
+      "eval_loss": 1.3529176712036133,
+      "eval_runtime": 52.82,
+      "eval_samples_per_second": 116.017,
+      "eval_steps_per_second": 3.635,
+      "step": 8000
+    },
+    {
+      "epoch": 0.4462645035963669,
+      "grad_norm": 1.1905977725982666,
+      "learning_rate": 2.7686774820181653e-05,
+      "loss": 1.4123,
+      "step": 8500
+    },
+    {
+      "epoch": 0.47251535674909434,
+      "grad_norm": 1.108475685119629,
+      "learning_rate": 2.637423216254528e-05,
+      "loss": 1.3987,
+      "step": 9000
+    },
+    {
+      "epoch": 0.47251535674909434,
+      "eval_accuracy": 0.6543280654604667,
+      "eval_loss": 1.3283472061157227,
+      "eval_runtime": 52.9443,
+      "eval_samples_per_second": 115.744,
+      "eval_steps_per_second": 3.626,
+      "step": 9000
+    },
+    {
+      "epoch": 0.4987662099018218,
+      "grad_norm": 1.173604965209961,
+      "learning_rate": 2.5061689504908907e-05,
+      "loss": 1.3864,
+      "step": 9500
+    },
+    {
+      "epoch": 0.5250170630545493,
+      "grad_norm": 1.187168836593628,
+      "learning_rate": 2.3749146847272537e-05,
+      "loss": 1.378,
+      "step": 10000
+    },
+    {
+      "epoch": 0.5250170630545493,
+      "eval_accuracy": 0.6578261027694616,
+      "eval_loss": 1.3098640441894531,
+      "eval_runtime": 52.7359,
+      "eval_samples_per_second": 116.202,
+      "eval_steps_per_second": 3.641,
+      "step": 10000
+    },
+    {
+      "epoch": 0.5512679162072768,
+      "grad_norm": 1.1469584703445435,
+      "learning_rate": 2.2436604189636164e-05,
+      "loss": 1.3679,
+      "step": 10500
+    },
+    {
+      "epoch": 0.5775187693600042,
+      "grad_norm": 1.1091519594192505,
+      "learning_rate": 2.112406153199979e-05,
+      "loss": 1.3567,
+      "step": 11000
+    },
+    {
+      "epoch": 0.5775187693600042,
+      "eval_accuracy": 0.6610596617229313,
+      "eval_loss": 1.2938392162322998,
+      "eval_runtime": 53.1407,
+      "eval_samples_per_second": 115.317,
+      "eval_steps_per_second": 3.613,
+      "step": 11000
+    },
+    {
+      "epoch": 0.6037696225127317,
+      "grad_norm": 1.1372675895690918,
+      "learning_rate": 1.9811518874363418e-05,
+      "loss": 1.3486,
+      "step": 11500
+    },
+    {
+      "epoch": 0.6300204756654592,
+      "grad_norm": 1.1002944707870483,
+      "learning_rate": 1.8498976216727045e-05,
+      "loss": 1.3436,
+      "step": 12000
+    },
+    {
+      "epoch": 0.6300204756654592,
+      "eval_accuracy": 0.6642933801929001,
+      "eval_loss": 1.2784197330474854,
+      "eval_runtime": 54.2879,
+      "eval_samples_per_second": 112.88,
+      "eval_steps_per_second": 3.537,
+      "step": 12000
+    },
+    {
+      "epoch": 0.6562713288181866,
+      "grad_norm": 1.1172466278076172,
+      "learning_rate": 1.718643355909067e-05,
+      "loss": 1.3349,
+      "step": 12500
+    },
+    {
+      "epoch": 0.6825221819709141,
+      "grad_norm": 1.129073143005371,
+      "learning_rate": 1.5873890901454296e-05,
+      "loss": 1.328,
+      "step": 13000
+    },
+    {
+      "epoch": 0.6825221819709141,
+      "eval_accuracy": 0.6665915343955856,
+      "eval_loss": 1.2663512229919434,
+      "eval_runtime": 52.6536,
+      "eval_samples_per_second": 116.383,
+      "eval_steps_per_second": 3.646,
+      "step": 13000
+    },
+    {
+      "epoch": 0.7087730351236415,
+      "grad_norm": 1.1130038499832153,
+      "learning_rate": 1.4561348243817924e-05,
+      "loss": 1.3209,
+      "step": 13500
+    },
+    {
+      "epoch": 0.735023888276369,
+      "grad_norm": 1.1710587739944458,
+      "learning_rate": 1.3248805586181551e-05,
+      "loss": 1.3157,
+      "step": 14000
+    },
+    {
+      "epoch": 0.735023888276369,
+      "eval_accuracy": 0.669028148919499,
+      "eval_loss": 1.2545733451843262,
+      "eval_runtime": 53.0376,
+      "eval_samples_per_second": 115.541,
+      "eval_steps_per_second": 3.62,
+      "step": 14000
+    },
+    {
+      "epoch": 0.7612747414290965,
+      "grad_norm": 1.1190043687820435,
+      "learning_rate": 1.1936262928545178e-05,
+      "loss": 1.3127,
+      "step": 14500
+    },
+    {
+      "epoch": 0.7875255945818239,
+      "grad_norm": 1.1405632495880127,
+      "learning_rate": 1.0623720270908805e-05,
+      "loss": 1.3052,
+      "step": 15000
+    },
+    {
+      "epoch": 0.7875255945818239,
+      "eval_accuracy": 0.6706863229277531,
+      "eval_loss": 1.2461665868759155,
+      "eval_runtime": 52.9855,
+      "eval_samples_per_second": 115.654,
+      "eval_steps_per_second": 3.624,
+      "step": 15000
+    },
+    {
+      "epoch": 0.8137764477345514,
+      "grad_norm": 1.159489393234253,
+      "learning_rate": 9.311177613272432e-06,
+      "loss": 1.3019,
+      "step": 15500
+    },
+    {
+      "epoch": 0.8400273008872788,
+      "grad_norm": 1.077958345413208,
+      "learning_rate": 7.998634955636059e-06,
+      "loss": 1.2966,
+      "step": 16000
+    },
+    {
+      "epoch": 0.8400273008872788,
+      "eval_accuracy": 0.6726187057979781,
+      "eval_loss": 1.2370483875274658,
+      "eval_runtime": 53.1969,
+      "eval_samples_per_second": 115.195,
+      "eval_steps_per_second": 3.609,
+      "step": 16000
+    },
+    {
+      "epoch": 0.8662781540400063,
+      "grad_norm": 1.1217402219772339,
+      "learning_rate": 6.686092297999686e-06,
+      "loss": 1.2936,
+      "step": 16500
+    },
+    {
+      "epoch": 0.8925290071927338,
+      "grad_norm": 1.1466121673583984,
+      "learning_rate": 5.373549640363313e-06,
+      "loss": 1.2943,
+      "step": 17000
+    },
+    {
+      "epoch": 0.8925290071927338,
+      "eval_accuracy": 0.6742480073198931,
+      "eval_loss": 1.2296273708343506,
+      "eval_runtime": 52.8045,
+      "eval_samples_per_second": 116.051,
+      "eval_steps_per_second": 3.636,
+      "step": 17000
+    },
+    {
+      "epoch": 0.9187798603454612,
+      "grad_norm": 1.1291422843933105,
+      "learning_rate": 4.061006982726939e-06,
+      "loss": 1.2877,
+      "step": 17500
+    },
+    {
+      "epoch": 0.9450307134981887,
+      "grad_norm": 1.1145981550216675,
+      "learning_rate": 2.7484643250905653e-06,
+      "loss": 1.2837,
+      "step": 18000
+    },
+    {
+      "epoch": 0.9450307134981887,
+      "eval_accuracy": 0.6753689297591429,
+      "eval_loss": 1.224404215812683,
+      "eval_runtime": 53.0184,
+      "eval_samples_per_second": 115.583,
+      "eval_steps_per_second": 3.621,
+      "step": 18000
+    },
+    {
+      "epoch": 0.9712815666509161,
+      "grad_norm": 1.0810720920562744,
+      "learning_rate": 1.4359216674541924e-06,
+      "loss": 1.2835,
+      "step": 18500
+    },
+    {
+      "epoch": 0.9975324198036436,
+      "grad_norm": 1.1252995729446411,
+      "learning_rate": 1.233790098178191e-07,
+      "loss": 1.2825,
+      "step": 19000
+    },
+    {
+      "epoch": 0.9975324198036436,
+      "eval_accuracy": 0.6759672761473065,
+      "eval_loss": 1.2216600179672241,
+      "eval_runtime": 52.7997,
+      "eval_samples_per_second": 116.061,
+      "eval_steps_per_second": 3.636,
+      "step": 19000
+    },
+    {
+      "epoch": 1.0,
+      "step": 19047,
+      "total_flos": 3.18502967574528e+17,
+      "train_loss": 1.5702329817773064,
+      "train_runtime": 7426.3864,
+      "train_samples_per_second": 82.069,
+      "train_steps_per_second": 2.565
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 19047,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.18502967574528e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf702862710b5bc311637f38f103bf1b146017dede9a874ef882b58c8aa0db0c
+size 5176