Duplicate from DavyMorgan/tiny-sd35-pipe

Co-authored-by: Yu Zheng <DavyMorgan@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,92 @@
+---
+library_name: diffusers
+---
+
+Pipeline generated with 
+
+```python
+import torch 
+from diffusers import AutoencoderKL, SD3Transformer2DModel, FlowMatchEulerDiscreteScheduler, StableDiffusion3Pipeline
+from transformers import CLIPTextConfig, CLIPTextModelWithProjection, T5EncoderModel, CLIPTokenizer, AutoTokenizer
+
+
+def get_dummy_components_sd3():
+    torch.manual_seed(0)
+    transformer = SD3Transformer2DModel(
+        sample_size=32,
+        patch_size=1,
+        in_channels=8,
+        num_layers=4,
+        attention_head_dim=8,
+        num_attention_heads=4,
+        joint_attention_dim=32,
+        caption_projection_dim=32,
+        pooled_projection_dim=64,
+        out_channels=8,
+        qk_norm="rms_norm",
+        dual_attention_layers=(0, 1),
+    )
+
+    torch.manual_seed(0)
+    clip_text_encoder_config = CLIPTextConfig(
+        bos_token_id=0,
+        eos_token_id=2,
+        hidden_size=32,
+        intermediate_size=37,
+        layer_norm_eps=1e-05,
+        num_attention_heads=4,
+        num_hidden_layers=5,
+        pad_token_id=1,
+        vocab_size=1000,
+        hidden_act="gelu",
+        projection_dim=32,
+    )
+
+    torch.manual_seed(0)
+    text_encoder = CLIPTextModelWithProjection(clip_text_encoder_config)
+
+    torch.manual_seed(0)
+    text_encoder_2 = CLIPTextModelWithProjection(clip_text_encoder_config)
+
+    torch.manual_seed(0)
+    text_encoder_3 = T5EncoderModel.from_pretrained("hf-internal-testing/tiny-random-t5")
+
+    tokenizer = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
+    tokenizer_2 = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
+    tokenizer_3 = AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-t5")
+
+    torch.manual_seed(0)
+    vae = AutoencoderKL(
+        sample_size=32,
+        in_channels=3,
+        out_channels=3,
+        block_out_channels=(4,),
+        layers_per_block=1,
+        latent_channels=8,
+        norm_num_groups=1,
+        use_quant_conv=False,
+        use_post_quant_conv=False,
+        shift_factor=0.0609,
+        scaling_factor=1.5035,
+    )
+
+    scheduler = FlowMatchEulerDiscreteScheduler()
+
+    return {
+        "scheduler": scheduler,
+        "text_encoder": text_encoder,
+        "text_encoder_2": text_encoder_2,
+        "text_encoder_3": text_encoder_3,
+        "tokenizer": tokenizer,
+        "tokenizer_2": tokenizer_2,
+        "tokenizer_3": tokenizer_3,
+        "transformer": transformer,
+        "vae": vae,
+    }
+
+
+if __name__ == "__main__":
+    components = get_dummy_components_sd3()
+    pipeline = StableDiffusion3Pipeline(**components)
+    pipeline.push_to_hub("DavyMorgan/tiny-sd35-pipe")
+```

model_index.json

@@ -0,0 +1,40 @@
+{
+  "_class_name": "StableDiffusion3Pipeline",
+  "_diffusers_version": "0.32.0.dev0",
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "CLIPTextModelWithProjection"
+  ],
+  "text_encoder_2": [
+    "transformers",
+    "CLIPTextModelWithProjection"
+  ],
+  "text_encoder_3": [
+    "transformers",
+    "T5EncoderModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "tokenizer_2": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "tokenizer_3": [
+    "transformers",
+    "T5TokenizerFast"
+  ],
+  "transformer": [
+    "diffusers",
+    "SD3Transformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,11 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.32.0.dev0",
+  "base_image_seq_len": 256,
+  "base_shift": 0.5,
+  "max_image_seq_len": 4096,
+  "max_shift": 1.15,
+  "num_train_timesteps": 1000,
+  "shift": 1.0,
+  "use_dynamic_shifting": false
+}

text_encoder/config.json

@@ -0,0 +1,23 @@
+{
+  "architectures": [
+    "CLIPTextModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_size": 32,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 37,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 5,
+  "pad_token_id": 1,
+  "projection_dim": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "vocab_size": 1000
+}

text_encoder/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:880e1efe0b6cefc228b92bf8f1acedaf3e6de60a950d38e61ca13210c2ef36b1
+size 287396

text_encoder_2/config.json

@@ -0,0 +1,23 @@
+{
+  "architectures": [
+    "CLIPTextModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_size": 32,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 37,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 5,
+  "pad_token_id": 1,
+  "projection_dim": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "vocab_size": 1000
+}

text_encoder_2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:880e1efe0b6cefc228b92bf8f1acedaf3e6de60a950d38e61ca13210c2ef36b1
+size 287396

text_encoder_3/config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "hf-internal-testing/tiny-random-t5",
+  "architectures": [
+    "T5EncoderModel"
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "d_ff": 37,
+  "d_kv": 8,
+  "d_model": 32,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "relu",
+  "gradient_checkpointing": false,
+  "initializer_factor": 0.002,
+  "is_encoder_decoder": true,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "num_decoder_layers": 5,
+  "num_heads": 4,
+  "num_layers": 5,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 8,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 1103
+}

text_encoder_3/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3ff1d3d80459481c9847d35de9a618835224139a4de455b40fcfe6f8544193b
+size 276696

tokenizer/merges.txt

@@ -0,0 +1,647 @@
+#version: 0.2
+Ġ t
+Ġt h
+Ġ a
+Ġth e</w>
+i n
+Ġ o
+Ġ ,</w>
+Ġ s
+e d</w>
+Ġ w
+e r
+Ġ .</w>
+Ġ i
+r e
+Ġ c
+n d</w>
+Ġ f
+Ġ b
+a t
+Ġo f</w>
+e r</w>
+e n
+a r
+o r
+i t
+Ġ p
+Ġ h
+Ġa nd</w>
+o n
+in g</w>
+a n
+r o
+Ġ m
+Ġ d
+e s</w>
+Ġi n</w>
+o n</w>
+Ġt o</w>
+o u
+i s
+Ġ a</w>
+i c
+Ġ T
+a l
+Ġ l
+Ġ =</w>
+Ġ re
+Ġ "</w>
+e s
+Ġ S
+a s</w>
+a l</w>
+i l
+e l
+i on</w>
+Ġ A
+Ġ C
+Ġ 1
+Ġ Ċ</w>
+u r
+ĠT h
+Ġ n
+a s
+Ġ @
+e c
+o m
+a c
+Ġ e
+Ġw as</w>
+Ġ M
+o r</w>
+a n</w>
+a m
+e n</w>
+o l
+Ġ in
+Ġ g
+Ġ '</w>
+Ġ B
+l y</w>
+a t</w>
+i v
+t s</w>
+ĠTh e</w>
+u s
+- @</w>
+Ġ@ -@</w>
+i s</w>
+Ġ I
+Ġw h
+i g
+Ġ H
+Ġs t
+o s
+u n
+t h
+Ġ P
+Ġw it
+Ġth at</w>
+i r
+Ġa s</w>
+e m
+Ġo n</w>
+r a
+Ġf or</w>
+Ġ R
+e t
+o w
+Ġ 2
+i d
+Ġ D
+l e</w>
+Ġwit h</w>
+l a
+en t</w>
+i m
+Ġ F
+e a
+i on
+Ġb y</w>
+Ġ )</w>
+Ġ (</w>
+Ġa l
+Ġc on
+en t
+Ġ W
+Ġi s</w>
+er e</w>
+Ġ G
+Ġ N
+Ġ L
+Ġh a
+er s</w>
+r i
+t h</w>
+t ed</w>
+u c
+Ġ J
+Ġ1 9
+e v
+u l
+Ġ v
+c e</w>
+at ion</w>
+ro m</w>
+Ġb e
+Ġ E
+i n</w>
+Ġth e
+Ġf rom</w>
+Ġ O
+t er</w>
+Ġp ro
+Ġa r
+a d
+Ġc om
+i c</w>
+a g
+Ġh is</w>
+Ġs h
+Ġa t</w>
+o v
+i es</w>
+o o
+p p
+s t
+c h
+Ġ r
+Ġ2 0
+a y</w>
+i f
+Ġw ere</w>
+Ġc h
+u t</w>
+s t</w>
+u t
+d s</w>
+o p
+u m
+Ġi t</w>
+o c
+t er
+l e
+ig h
+u d
+Ġe x
+ion s</w>
+at e</w>
+it y</w>
+at ed</w>
+Ġ un
+e p
+q u
+Ġn o
+Ġ K
+iv e</w>
+is t
+Ġo n
+am e</w>
+ou n
+i r</w>
+a b
+Ġ â
+in g
+Ġh e</w>
+l d</w>
+u g
+ic h</w>
+Ġa n</w>
+e d
+Ġ k
+Ġâ Ģ
+Ġha d</w>
+v e</w>
+a in
+Ġs e
+t ion</w>
+or e</w>
+re s
+Ġwh ich</w>
+ĠI n</w>
+o d
+th er</w>
+a k
+Ġs p
+a r</w>
+Ġ y
+ĠC h
+on g</w>
+Ġa c
+es t</w>
+Ġ U
+a p
+f f
+al ly</w>
+r it
+ĠS t
+u b
+g e</w>
+b er</w>
+e t</w>
+Ġb e</w>
+e ar
+Ġre c
+er s
+Ġf ir
+o t
+Ġar e</w>
+Ġa n
+c h</w>
+o g
+i a</w>
+es t
+in e</w>
+il l
+an d
+e l</w>
+ar y</w>
+e w</w>
+i d</w>
+Ġf or
+Ġ ;</w>
+Ġcom p
+Ġ V
+Ġin c
+t r
+Ġ20 0
+Ġthe ir</w>
+u s</w>
+Ġb ut</w>
+r an
+ic al</w>
+Ġfir st</w>
+Ġd e
+Ġin t
+Ġ ro
+s o</w>
+ĠâĢ ĵ</w>
+Ġno t</w>
+d ing</w>
+f ter</w>
+ur e</w>
+Ġp ar
+Ġ :</w>
+i an</w>
+Ġt w
+ou ld</w>
+Ġal so</w>
+Ġi ts</w>
+Ġw or
+u m</w>
+Ġo r</w>
+os t</w>
+0 0</w>
+ou r
+ar d</w>
+Ġre s
+m p
+u e</w>
+Ġa b
+is h</w>
+Ġcon t
+Ġa d
+ow n</w>
+al l</w>
+ou g
+Ġh er</w>
+as t</w>
+Ġ en
+om e</w>
+al l
+d ed</w>
+o w</w>
+Ġha ve</w>
+Ġ us
+ea r</w>
+ac k</w>
+d uc
+i al</w>
+s s
+en ts</w>
+a in</w>
+t ing</w>
+Ġon e</w>
+es s
+Ġh as</w>
+igh t</w>
+a v
+Ġe v
+ou t</w>
+a y
+en ce</w>
+Ġbe en</w>
+e w
+Ġtw o</w>
+Ġc l
+d er</w>
+im e</w>
+k s</w>
+es s</w>
+is h
+. @</w>
+Ġ@ .@</w>
+Ġp la
+Ġp l
+Ġo r
+u p</w>
+m ent</w>
+ur ing</w>
+ol l
+ĠI n
+Ġth is</w>
+Ġb ec
+Ġcom m
+Ġd is
+at er</w>
+ag e</w>
+Ġa pp
+ou s</w>
+e y</w>
+i l</w>
+p er
+ĠA l
+ion al</w>
+l ud
+el y</w>
+t t
+il e</w>
+i z
+Ġ j
+Ġwh o</w>
+Ġa g
+i b
+Ġthe y</w>
+f or
+Ġo v
+at h
+e g
+Ġs c
+i p
+Ġ20 1
+Ġ 3
+Ġp er
+or y</w>
+Ġd es
+id e</w>
+Ġs er
+s e</w>
+ĠH e</w>
+la nd</w>
+at ions</w>
+r ic
+i t</w>
+re s</w>
+er ed</w>
+Ġp re
+ĠS h
+an ce</w>
+or t</w>
+an t</w>
+, @</w>
+Ġ@ ,@</w>
+el l</w>
+Ġ Y
+n ed</w>
+el l
+it e</w>
+Ġinc lud
+Ġre p
+Ġa fter</w>
+Ġs uc
+re e</w>
+an y</w>
+i m</w>
+or t
+Ġ1 8
+Ġs u
+ad e</w>
+ou r</w>
+ĠU n
+ĠI t</w>
+i k
+ĠM ar
+em ber</w>
+Ġ 1</w>
+e en</w>
+a nd</w>
+Ġs ec
+ic e</w>
+Ġt ime</w>
+ĠA n
+Ġint o</w>
+Ġf in
+Ġo ther</w>
+Ġa tt
+il l</w>
+re n
+ac h
+as s
+er al</w>
+es e</w>
+s h
+al s</w>
+it ion</w>
+oug h</w>
+l es</w>
+am p
+Ġw ould</w>
+Ġm ore</w>
+ro ug
+ri b
+er y</w>
+ac e</w>
+Ġ A</w>
+Ġpla y
+it ed</w>
+k ed</w>
+is t</w>
+i ed</w>
+Ġ 2</w>
+as ed</w>
+ing s</w>
+an g
+a m</w>
+i p</w>
+Ġb o
+ab le</w>
+t y</w>
+Ġch ar
+Ġc ent
+et w
+at es</w>
+ro p
+Ġ I</w>
+u nd</w>
+ĠA m
+c es</w>
+o in
+Ġin ter
+u p
+c t
+on e</w>
+Ġt ra
+an t
+ec t
+Ġal l</w>
+e f
+Ġcon s
+ub l
+n ing</w>
+an s</w>
+Ġf e
+us t</w>
+Ġ 0
+Ġre m
+as e</w>
+on g
+Ġwh en</w>
+e b
+ĠW h
+Ġe ar
+ev er</w>
+Ġov er</w>
+Ġk n
+a us
+Ġp os
+a d</w>
+er m
+Ġsh e</w>
+Ġ ra
+Ġd uring</w>
+as on</w>
+v i
+Ġex p
+Ġl ea
+Ġ el
+Ġ 4
+Ġon ly</w>
+o nd</w>
+Ġd ec
+Ġac c
+Ġo ff
+is s
+Ġf l
+ĠE n
+o t</w>
+en s
+os e</w>
+ak e</w>
+o m</w>
+Ġs ev
+ac h</w>
+etw een</w>
+er n
+Ġ 3</w>
+Ġp r
+Ġg ro
+r uc
+Ġd i
+Ġ19 9
+ĠA r
+Ġg ame</w>
+Ġh im</w>
+oo k</w>
+Ġ up</w>
+Ġab out</w>
+Ġre l
+for m
+Ġth ree</w>
+at t
+ĠC om
+Ġs a
+ear s</w>
+Ġ 5
+r y</w>
+Ġi mp
+Ġm ost</w>
+f er
+Ġp res
+Ġf il
+Ġb etween</w>
+Ġbe g
+p h
+or s</w>
+Ġth an</w>
+Ġrec or
+o b
+er ic
+at ing</w>
+Ġth roug
+k ing</w>
+Ġo ut</w>
+Ġn um
+oo d</w>
+oll ow
+ac t
+u il
+Ġc re
+ol og
+at ional</w>
+Ġpro duc
+Ġwh ile</w>
+Ġl ater</w>
+Ġw rit
+e x
+Ġst ar
+Ġsp ec
+e e
+ish ed</w>
+Ġre g
+is ion</w>
+ou th</w>
+Ġre le
+Ġa ss
+Ġse ason</w>
+Ġm ade</w>
+il y</w>
+r u
+o y
+t ur
+t e</w>
+Ġ qu
+Ġm ov
+ur y</w>
+ĠAm eric
+em ent</w>
+c c
+ou nd</w>
+Ġl ar
+Ġfor m
+ec t</w>
+Ġde f
+Ġm us
+ĠP ar
+Ġm e
+Ġs ub
+w ay</w>
+o p</w>
+o h
+el d</w>
+i e</w>
+em p
+am es</w>
+er n</w>
+Ġn or
+iv ed</w>
+ev el
+Ġsuc h</w>
+ar ds</w>
+Ġin d
+ik e</w>
+Ġg en
+er t
+Ġy ear</w>
+Ġus ed</w>
+Ġn ew</w>
+Ġ 5</w>
+Ġal b
+s p
+y p
+Ġwit h
+Ġwh ere</w>
+ic s</w>
+ĠTh is</w>
+Ġthe m</w>
+w n</w>

tokenizer/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,30 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|startoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 77,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": "<|endoftext|>"
+}

tokenizer/vocab.json

@@ -0,0 +1,1002 @@
+{
+  "!": 2,
+  "!</w>": 345,
+  "\"": 3,
+  "\"</w>": 344,
+  "#": 4,
+  "#</w>": 325,
+  "$": 5,
+  "$</w>": 348,
+  "%": 6,
+  "%</w>": 351,
+  "&": 7,
+  "&</w>": 352,
+  "'": 8,
+  "'</w>": 296,
+  "(": 9,
+  "(</w>": 318,
+  ")": 10,
+  ")</w>": 330,
+  "*": 11,
+  "*</w>": 327,
+  "+": 12,
+  "+</w>": 341,
+  ",": 13,
+  ",</w>": 279,
+  ",@</w>": 754,
+  "-": 14,
+  "-</w>": 276,
+  "-@</w>": 439,
+  ".": 15,
+  ".</w>": 253,
+  ".@</w>": 695,
+  "/": 16,
+  "/</w>": 350,
+  "0": 17,
+  "00</w>": 647,
+  "0</w>": 216,
+  "1": 18,
+  "1</w>": 222,
+  "2": 19,
+  "2</w>": 231,
+  "3": 20,
+  "3</w>": 243,
+  "4": 21,
+  "4</w>": 233,
+  "5": 22,
+  "5</w>": 240,
+  "6": 23,
+  "6</w>": 226,
+  "7": 24,
+  "7</w>": 215,
+  "8": 25,
+  "8</w>": 236,
+  "9": 26,
+  "9</w>": 242,
+  ":": 27,
+  ":</w>": 353,
+  ";": 28,
+  ";</w>": 317,
+  "<": 29,
+  "<</w>": 340,
+  "<|endoftext|>": 1,
+  "<|startoftext|>": 0,
+  "=": 30,
+  "=</w>": 342,
+  ">": 31,
+  "></w>": 300,
+  "?": 32,
+  "?</w>": 346,
+  "@": 33,
+  "@</w>": 320,
+  "A": 34,
+  "A</w>": 227,
+  "B": 35,
+  "B</w>": 258,
+  "C": 36,
+  "C</w>": 239,
+  "D": 37,
+  "D</w>": 255,
+  "E": 38,
+  "E</w>": 246,
+  "F": 39,
+  "F</w>": 213,
+  "G": 40,
+  "G</w>": 283,
+  "H": 41,
+  "H</w>": 219,
+  "I": 42,
+  "I</w>": 237,
+  "J": 43,
+  "J</w>": 251,
+  "K": 44,
+  "K</w>": 254,
+  "L": 45,
+  "L</w>": 218,
+  "M": 46,
+  "M</w>": 234,
+  "N": 47,
+  "N</w>": 238,
+  "O": 48,
+  "O</w>": 265,
+  "P": 49,
+  "P</w>": 245,
+  "Q": 50,
+  "Q</w>": 309,
+  "R": 51,
+  "R</w>": 264,
+  "S": 52,
+  "S</w>": 230,
+  "T": 53,
+  "T</w>": 235,
+  "U": 54,
+  "U</w>": 268,
+  "V": 55,
+  "V</w>": 248,
+  "W": 56,
+  "W</w>": 274,
+  "X": 57,
+  "X</w>": 263,
+  "Y": 58,
+  "Y</w>": 310,
+  "Z": 59,
+  "Z</w>": 207,
+  "[": 60,
+  "[</w>": 270,
+  "\\": 61,
+  "\\</w>": 338,
+  "]": 62,
+  "]</w>": 289,
+  "^": 63,
+  "^</w>": 331,
+  "_": 64,
+  "_</w>": 334,
+  "`": 65,
+  "`</w>": 347,
+  "a": 66,
+  "a</w>": 197,
+  "ab": 555,
+  "able</w>": 820,
+  "ac": 420,
+  "ace</w>": 806,
+  "ach": 791,
+  "ach</w>": 885,
+  "ack</w>": 670,
+  "act": 929,
+  "ad": 508,
+  "ad</w>": 860,
+  "ade</w>": 771,
+  "ag": 511,
+  "age</w>": 710,
+  "ain": 568,
+  "ain</w>": 675,
+  "ak": 577,
+  "ake</w>": 882,
+  "al": 397,
+  "al</w>": 405,
+  "all": 664,
+  "all</w>": 658,
+  "ally</w>": 588,
+  "als</w>": 796,
+  "am": 426,
+  "am</w>": 817,
+  "ame</w>": 552,
+  "ames</w>": 976,
+  "amp": 800,
+  "an": 384,
+  "an</w>": 425,
+  "ance</w>": 751,
+  "and": 609,
+  "and</w>": 780,
+  "ang": 816,
+  "ans</w>": 844,
+  "ant": 837,
+  "ant</w>": 753,
+  "any</w>": 766,
+  "ap": 586,
+  "ar": 376,
+  "ar</w>": 579,
+  "ard</w>": 649,
+  "ards</w>": 982,
+  "ary</w>": 611,
+  "as": 416,
+  "as</w>": 404,
+  "ase</w>": 849,
+  "ased</w>": 814,
+  "ason</w>": 865,
+  "ass": 792,
+  "ast</w>": 661,
+  "at": 372,
+  "at</w>": 434,
+  "ate</w>": 541,
+  "ated</w>": 543,
+  "ater</w>": 709,
+  "ates</w>": 825,
+  "ath": 730,
+  "ating</w>": 922,
+  "ation</w>": 497,
+  "ational</w>": 933,
+  "ations</w>": 744,
+  "att": 903,
+  "aus": 858,
+  "av": 681,
+  "ay": 684,
+  "ay</w>": 523,
+  "b": 67,
+  "b</w>": 212,
+  "ber</w>": 593,
+  "c": 68,
+  "c</w>": 224,
+  "cc": 960,
+  "ce</w>": 496,
+  "ces</w>": 830,
+  "ch": 520,
+  "ch</w>": 603,
+  "ct": 834,
+  "d": 69,
+  "d</w>": 196,
+  "ded</w>": 665,
+  "der</w>": 690,
+  "ding</w>": 633,
+  "ds</w>": 530,
+  "duc": 671,
+  "e": 70,
+  "e</w>": 195,
+  "ea": 471,
+  "ear": 596,
+  "ear</w>": 669,
+  "ears</w>": 906,
+  "eb": 852,
+  "ec": 418,
+  "ect": 838,
+  "ect</w>": 964,
+  "ed": 563,
+  "ed</w>": 362,
+  "ee": 941,
+  "een</w>": 779,
+  "ef": 840,
+  "eg": 731,
+  "el": 407,
+  "el</w>": 610,
+  "eld</w>": 973,
+  "ell": 759,
+  "ell</w>": 756,
+  "ely</w>": 719,
+  "em": 455,
+  "ember</w>": 777,
+  "ement</w>": 959,
+  "emp": 975,
+  "en": 375,
+  "en</w>": 427,
+  "ence</w>": 685,
+  "ens": 880,
+  "ent": 478,
+  "ent</w>": 468,
+  "ents</w>": 674,
+  "ep": 545,
+  "er": 364,
+  "er</w>": 374,
+  "eral</w>": 793,
+  "ere</w>": 481,
+  "ered</w>": 748,
+  "eric": 921,
+  "erm": 861,
+  "ern": 887,
+  "ern</w>": 977,
+  "ers": 598,
+  "ers</w>": 486,
+  "ert": 986,
+  "ery</w>": 805,
+  "es": 402,
+  "es</w>": 388,
+  "ese</w>": 794,
+  "ess": 678,
+  "ess</w>": 693,
+  "est": 606,
+  "est</w>": 584,
+  "et": 460,
+  "et</w>": 594,
+  "etw": 824,
+  "etween</w>": 886,
+  "ev": 493,
+  "evel": 980,
+  "ever</w>": 855,
+  "ew": 687,
+  "ew</w>": 612,
+  "ex": 938,
+  "ey</w>": 713,
+  "f": 71,
+  "f</w>": 209,
+  "fer": 911,
+  "ff": 587,
+  "for": 728,
+  "form": 901,
+  "fter</w>": 634,
+  "g": 72,
+  "g</w>": 214,
+  "ge</w>": 592,
+  "h": 73,
+  "h</w>": 203,
+  "i": 74,
+  "i</w>": 205,
+  "ia</w>": 605,
+  "ial</w>": 672,
+  "ian</w>": 638,
+  "ib": 726,
+  "ic": 395,
+  "ic</w>": 510,
+  "ical</w>": 625,
+  "ice</w>": 782,
+  "ich</w>": 561,
+  "ics</w>": 996,
+  "id": 463,
+  "id</w>": 613,
+  "ide</w>": 739,
+  "ie</w>": 974,
+  "ied</w>": 812,
+  "ies</w>": 516,
+  "if": 524,
+  "ig": 444,
+  "igh": 537,
+  "ight</w>": 680,
+  "ik": 775,
+  "ike</w>": 984,
+  "il": 406,
+  "il</w>": 714,
+  "ile</w>": 721,
+  "ill": 608,
+  "ill</w>": 789,
+  "ily</w>": 950,
+  "im": 469,
+  "im</w>": 767,
+  "ime</w>": 691,
+  "in": 358,
+  "in</w>": 501,
+  "ine</w>": 607,
+  "ing": 557,
+  "ing</w>": 383,
+  "ings</w>": 815,
+  "ion": 472,
+  "ion</w>": 408,
+  "ional</w>": 717,
+  "ions</w>": 540,
+  "ip": 733,
+  "ip</w>": 818,
+  "ir": 453,
+  "ir</w>": 554,
+  "is": 393,
+  "is</w>": 441,
+  "ish": 694,
+  "ish</w>": 654,
+  "ished</w>": 942,
+  "ision</w>": 944,
+  "iss": 876,
+  "ist": 550,
+  "ist</w>": 811,
+  "it": 378,
+  "it</w>": 746,
+  "ite</w>": 760,
+  "ited</w>": 809,
+  "ition</w>": 797,
+  "ity</w>": 542,
+  "iv": 435,
+  "ive</w>": 549,
+  "ived</w>": 979,
+  "iz": 722,
+  "j": 75,
+  "j</w>": 288,
+  "k": 76,
+  "k</w>": 210,
+  "ked</w>": 810,
+  "king</w>": 924,
+  "ks</w>": 692,
+  "l": 77,
+  "l</w>": 201,
+  "la": 467,
+  "land</w>": 743,
+  "ld</w>": 559,
+  "le": 536,
+  "le</w>": 465,
+  "les</w>": 799,
+  "lud": 718,
+  "ly</w>": 433,
+  "m": 78,
+  "m</w>": 202,
+  "ment</w>": 701,
+  "mp": 651,
+  "n": 79,
+  "n</w>": 199,
+  "nd</w>": 369,
+  "ned</w>": 758,
+  "ning</w>": 843,
+  "o": 80,
+  "o</w>": 198,
+  "ob": 920,
+  "oc": 534,
+  "od": 575,
+  "og": 604,
+  "oh": 972,
+  "oin": 831,
+  "ol": 428,
+  "oll": 703,
+  "ollow": 928,
+  "olog": 932,
+  "om": 419,
+  "om</w>": 883,
+  "ome</w>": 663,
+  "on": 382,
+  "on</w>": 390,
+  "ond</w>": 872,
+  "one</w>": 835,
+  "ong": 850,
+  "ong</w>": 582,
+  "oo": 517,
+  "ood</w>": 927,
+  "ook</w>": 897,
+  "op": 531,
+  "op</w>": 971,
+  "or": 377,
+  "or</w>": 424,
+  "ore</w>": 571,
+  "ors</w>": 917,
+  "ort": 768,
+  "ort</w>": 752,
+  "ory</w>": 737,
+  "os": 447,
+  "ose</w>": 881,
+  "ost</w>": 646,
+  "ot": 600,
+  "ot</w>": 879,
+  "ou": 392,
+  "oug": 659,
+  "ough</w>": 798,
+  "ould</w>": 640,
+  "oun": 553,
+  "ound</w>": 961,
+  "our": 648,
+  "our</w>": 772,
+  "ous</w>": 712,
+  "out</w>": 683,
+  "outh</w>": 945,
+  "ov": 515,
+  "ow": 461,
+  "ow</w>": 666,
+  "own</w>": 657,
+  "oy": 952,
+  "p": 81,
+  "p</w>": 217,
+  "per": 715,
+  "ph": 916,
+  "pp": 518,
+  "q": 82,
+  "q</w>": 280,
+  "qu": 546,
+  "r": 83,
+  "r</w>": 204,
+  "ra": 457,
+  "ran": 624,
+  "re": 367,
+  "ree</w>": 765,
+  "ren": 790,
+  "res": 572,
+  "res</w>": 747,
+  "ri": 487,
+  "rib": 804,
+  "ric": 745,
+  "rit": 589,
+  "ro": 385,
+  "rom</w>": 498,
+  "rop": 826,
+  "roug": 803,
+  "ru": 951,
+  "ruc": 891,
+  "ry</w>": 908,
+  "s": 84,
+  "s</w>": 206,
+  "se</w>": 741,
+  "sh": 795,
+  "so</w>": 630,
+  "sp": 992,
+  "ss": 673,
+  "st": 519,
+  "st</w>": 528,
+  "t": 85,
+  "t</w>": 208,
+  "te</w>": 954,
+  "ted</w>": 489,
+  "ter": 535,
+  "ter</w>": 505,
+  "th": 449,
+  "th</w>": 488,
+  "ther</w>": 576,
+  "ting</w>": 676,
+  "tion</w>": 570,
+  "tr": 619,
+  "ts</w>": 436,
+  "tt": 720,
+  "tur": 953,
+  "ty</w>": 821,
+  "u": 86,
+  "u</w>": 229,
+  "ub": 591,
+  "ubl": 842,
+  "uc": 490,
+  "ud": 538,
+  "ue</w>": 652,
+  "ug": 560,
+  "uil": 930,
+  "ul": 494,
+  "um": 532,
+  "um</w>": 644,
+  "un": 448,
+  "und</w>": 828,
+  "up": 833,
+  "up</w>": 700,
+  "ur": 413,
+  "ure</w>": 635,
+  "uring</w>": 702,
+  "ury</w>": 957,
+  "us": 438,
+  "us</w>": 622,
+  "ust</w>": 846,
+  "ut": 529,
+  "ut</w>": 527,
+  "v": 87,
+  "v</w>": 232,
+  "ve</w>": 567,
+  "vi": 866,
+  "w": 88,
+  "w</w>": 250,
+  "way</w>": 970,
+  "wn</w>": 999,
+  "x": 89,
+  "x</w>": 269,
+  "y": 90,
+  "y</w>": 211,
+  "yp": 993,
+  "z": 91,
+  "z</w>": 228,
+  "|": 92,
+  "|</w>": 304,
+  "}": 93,
+  "}</w>": 336,
+  "~": 94,
+  "~</w>": 343,
+  "¡": 95,
+  "¡</w>": 220,
+  "¢": 96,
+  "¢</w>": 306,
+  "£": 97,
+  "£</w>": 323,
+  "¤": 98,
+  "¤</w>": 292,
+  "¥": 99,
+  "¥</w>": 339,
+  "¦": 100,
+  "¦</w>": 303,
+  "§": 101,
+  "§</w>": 275,
+  "¨": 102,
+  "¨</w>": 282,
+  "©": 103,
+  "©</w>": 259,
+  "ª": 104,
+  "ª</w>": 286,
+  "«": 105,
+  "«</w>": 266,
+  "¬": 106,
+  "¬</w>": 319,
+  "®": 107,
+  "®</w>": 329,
+  "¯": 108,
+  "¯</w>": 287,
+  "°": 109,
+  "°</w>": 298,
+  "±": 110,
+  "±</w>": 200,
+  "²": 111,
+  "²</w>": 284,
+  "³": 112,
+  "³</w>": 272,
+  "´": 113,
+  "´</w>": 307,
+  "µ": 114,
+  "µ</w>": 261,
+  "¶": 115,
+  "¶</w>": 301,
+  "·": 116,
+  "·</w>": 326,
+  "¸": 117,
+  "¸</w>": 257,
+  "¹": 118,
+  "¹</w>": 241,
+  "º": 119,
+  "º</w>": 260,
+  "»": 120,
+  "»</w>": 247,
+  "¼": 121,
+  "¼</w>": 305,
+  "½": 122,
+  "½</w>": 294,
+  "¾": 123,
+  "¾</w>": 316,
+  "¿": 124,
+  "¿</w>": 271,
+  "Â": 125,
+  "Ã": 126,
+  "Ä": 127,
+  "Å": 128,
+  "Æ": 129,
+  "Ç": 130,
+  "È": 131,
+  "É": 132,
+  "Ê": 133,
+  "Ë": 134,
+  "Ì": 135,
+  "Í": 136,
+  "Î": 137,
+  "Ï": 138,
+  "Ð": 139,
+  "Ñ": 140,
+  "Ö": 141,
+  "×": 142,
+  "Ø": 143,
+  "Ù": 144,
+  "Ü": 145,
+  "à": 146,
+  "á": 147,
+  "â": 148,
+  "ã": 149,
+  "ä": 150,
+  "å": 151,
+  "æ": 152,
+  "ç": 153,
+  "è": 154,
+  "é": 155,
+  "ë": 156,
+  "ì": 157,
+  "ï": 158,
+  "Ċ": 159,
+  "Ċ</w>": 349,
+  "Ġ": 160,
+  "Ġ\"</w>": 401,
+  "Ġ'</w>": 431,
+  "Ġ(</w>": 475,
+  "Ġ)</w>": 474,
+  "Ġ,</w>": 360,
+  "Ġ.</w>": 365,
+  "Ġ0": 847,
+  "Ġ1": 411,
+  "Ġ18": 769,
+  "Ġ19": 492,
+  "Ġ199": 893,
+  "Ġ1</w>": 778,
+  "Ġ2": 462,
+  "Ġ20": 522,
+  "Ġ200": 620,
+  "Ġ201": 734,
+  "Ġ2</w>": 813,
+  "Ġ3": 735,
+  "Ġ3</w>": 888,
+  "Ġ4": 870,
+  "Ġ5": 907,
+  "Ġ5</w>": 990,
+  "Ġ:</w>": 637,
+  "Ġ;</w>": 615,
+  "Ġ</w>": 333,
+  "Ġ=</w>": 399,
+  "Ġ@": 417,
+  "Ġ@,@</w>": 755,
+  "Ġ@-@</w>": 440,
+  "Ġ@.@</w>": 696,
+  "ĠA": 409,
+  "ĠA</w>": 807,
+  "ĠAl": 716,
+  "ĠAm": 829,
+  "ĠAmeric": 958,
+  "ĠAn": 784,
+  "ĠAr": 894,
+  "ĠB": 432,
+  "ĠC": 410,
+  "ĠCh": 581,
+  "ĠCom": 904,
+  "ĠD": 464,
+  "ĠE": 500,
+  "ĠEn": 878,
+  "ĠF": 470,
+  "ĠG": 482,
+  "ĠH": 445,
+  "ĠHe</w>": 742,
+  "ĠI": 442,
+  "ĠI</w>": 827,
+  "ĠIn": 704,
+  "ĠIn</w>": 574,
+  "ĠIt</w>": 774,
+  "ĠJ": 491,
+  "ĠK": 548,
+  "ĠL": 484,
+  "ĠM": 423,
+  "ĠMar": 776,
+  "ĠN": 483,
+  "ĠO": 504,
+  "ĠP": 450,
+  "ĠPar": 967,
+  "ĠR": 459,
+  "ĠS": 403,
+  "ĠSh": 750,
+  "ĠSt": 590,
+  "ĠT": 396,
+  "ĠTh": 414,
+  "ĠThe</w>": 437,
+  "ĠThis</w>": 997,
+  "ĠU": 585,
+  "ĠUn": 773,
+  "ĠV": 617,
+  "ĠW": 479,
+  "ĠWh": 853,
+  "ĠY": 757,
+  "Ġa": 356,
+  "Ġa</w>": 394,
+  "Ġab": 653,
+  "Ġabout</w>": 899,
+  "Ġac": 583,
+  "Ġacc": 874,
+  "Ġad": 656,
+  "Ġafter</w>": 763,
+  "Ġag": 725,
+  "Ġal": 476,
+  "Ġalb": 991,
+  "Ġall</w>": 839,
+  "Ġalso</w>": 641,
+  "Ġan": 602,
+  "Ġan</w>": 562,
+  "Ġand</w>": 381,
+  "Ġapp": 711,
+  "Ġar": 507,
+  "Ġare</w>": 601,
+  "Ġas</w>": 454,
+  "Ġass": 947,
+  "Ġat</w>": 514,
+  "Ġatt": 788,
+  "Ġb": 371,
+  "Ġbe": 499,
+  "Ġbe</w>": 595,
+  "Ġbec": 706,
+  "Ġbeen</w>": 686,
+  "Ġbeg": 915,
+  "Ġbetween</w>": 914,
+  "Ġbo": 819,
+  "Ġbut</w>": 623,
+  "Ġby</w>": 473,
+  "Ġc": 368,
+  "Ġcent": 823,
+  "Ġch": 526,
+  "Ġchar": 822,
+  "Ġcl": 689,
+  "Ġcom": 509,
+  "Ġcomm": 707,
+  "Ġcomp": 616,
+  "Ġcon": 477,
+  "Ġcons": 841,
+  "Ġcont": 655,
+  "Ġcre": 931,
+  "Ġd": 387,
+  "Ġde": 627,
+  "Ġdec": 873,
+  "Ġdef": 965,
+  "Ġdes": 738,
+  "Ġdi": 892,
+  "Ġdis": 708,
+  "Ġduring</w>": 864,
+  "Ġe": 421,
+  "Ġear": 854,
+  "Ġel": 869,
+  "Ġen": 662,
+  "Ġev": 682,
+  "Ġex": 539,
+  "Ġexp": 867,
+  "Ġf": 370,
+  "Ġfe": 845,
+  "Ġfil": 913,
+  "Ġfin": 786,
+  "Ġfir": 599,
+  "Ġfirst</w>": 626,
+  "Ġfl": 877,
+  "Ġfor": 614,
+  "Ġfor</w>": 458,
+  "Ġform": 963,
+  "Ġfrom</w>": 503,
+  "Ġg": 430,
+  "Ġgame</w>": 895,
+  "Ġgen": 985,
+  "Ġgro": 890,
+  "Ġh": 380,
+  "Ġha": 485,
+  "Ġhad</w>": 566,
+  "Ġhas</w>": 679,
+  "Ġhave</w>": 667,
+  "Ġhe</w>": 558,
+  "Ġher</w>": 660,
+  "Ġhim</w>": 896,
+  "Ġhis</w>": 512,
+  "Ġi": 366,
+  "Ġimp": 909,
+  "Ġin": 429,
+  "Ġin</w>": 389,
+  "Ġinc": 618,
+  "Ġinclud": 761,
+  "Ġind": 983,
+  "Ġint": 628,
+  "Ġinter": 832,
+  "Ġinto</w>": 785,
+  "Ġis</w>": 480,
+  "Ġit</w>": 533,
+  "Ġits</w>": 642,
+  "Ġj": 723,
+  "Ġk": 564,
+  "Ġkn": 857,
+  "Ġl": 398,
+  "Ġlar": 962,
+  "Ġlater</w>": 936,
+  "Ġlea": 868,
+  "Ġm": 386,
+  "Ġmade</w>": 949,
+  "Ġme": 968,
+  "Ġmore</w>": 802,
+  "Ġmost</w>": 910,
+  "Ġmov": 956,
+  "Ġmus": 966,
+  "Ġn": 415,
+  "Ġnew</w>": 989,
+  "Ġno": 547,
+  "Ġnor": 978,
+  "Ġnot</w>": 632,
+  "Ġnum": 926,
+  "Ġo": 359,
+  "Ġof</w>": 373,
+  "Ġoff": 875,
+  "Ġon": 551,
+  "Ġon</w>": 456,
+  "Ġone</w>": 677,
+  "Ġonly</w>": 871,
+  "Ġor": 699,
+  "Ġor</w>": 645,
+  "Ġother</w>": 787,
+  "Ġout</w>": 925,
+  "Ġov": 729,
+  "Ġover</w>": 856,
+  "Ġp": 379,
+  "Ġpar": 636,
+  "Ġper": 736,
+  "Ġpl": 698,
+  "Ġpla": 697,
+  "Ġplay": 808,
+  "Ġpos": 859,
+  "Ġpr": 889,
+  "Ġpre": 749,
+  "Ġpres": 912,
+  "Ġpro": 506,
+  "Ġproduc": 934,
+  "Ġqu": 955,
+  "Ġr": 521,
+  "Ġra": 863,
+  "Ġre": 400,
+  "Ġrec": 597,
+  "Ġrecor": 919,
+  "Ġreg": 943,
+  "Ġrel": 900,
+  "Ġrele": 946,
+  "Ġrem": 848,
+  "Ġrep": 762,
+  "Ġres": 650,
+  "Ġro": 629,
+  "Ġs": 361,
+  "Ġsa": 905,
+  "Ġsc": 732,
+  "Ġse": 569,
+  "Ġseason</w>": 948,
+  "Ġsec": 781,
+  "Ġser": 740,
+  "Ġsev": 884,
+  "Ġsh": 513,
+  "Ġshe</w>": 862,
+  "Ġsp": 578,
+  "Ġspec": 940,
+  "Ġst": 446,
+  "Ġstar": 939,
+  "Ġsu": 770,
+  "Ġsub": 969,
+  "Ġsuc": 764,
+  "Ġsuch</w>": 981,
+  "Ġt": 354,
+  "Ġth": 355,
+  "Ġthan</w>": 918,
+  "Ġthat</w>": 452,
+  "Ġthe": 502,
+  "Ġthe</w>": 357,
+  "Ġtheir</w>": 621,
+  "Ġthem</w>": 998,
+  "Ġthey</w>": 727,
+  "Ġthis</w>": 705,
+  "Ġthree</w>": 902,
+  "Ġthroug": 923,
+  "Ġtime</w>": 783,
+  "Ġto</w>": 391,
+  "Ġtra": 836,
+  "Ġtw": 639,
+  "Ġtwo</w>": 688,
+  "Ġun": 544,
+  "Ġup</w>": 898,
+  "Ġus": 668,
+  "Ġused</w>": 988,
+  "Ġv": 495,
+  "Ġw": 363,
+  "Ġwas</w>": 422,
+  "Ġwere</w>": 525,
+  "Ġwh": 443,
+  "Ġwhen</w>": 851,
+  "Ġwhere</w>": 995,
+  "Ġwhich</w>": 573,
+  "Ġwhile</w>": 935,
+  "Ġwho</w>": 724,
+  "Ġwit": 451,
+  "Ġwith": 994,
+  "Ġwith</w>": 466,
+  "Ġwor": 643,
+  "Ġwould</w>": 801,
+  "Ġwrit": 937,
+  "Ġy": 580,
+  "Ġyear</w>": 987,
+  "Ġâ": 556,
+  "ĠâĢ": 565,
+  "ĠâĢĵ</w>": 631,
+  "ĠĊ</w>": 412,
+  "Ģ": 161,
+  "Ģ</w>": 223,
+  "ģ": 162,
+  "ģ</w>": 273,
+  "Ĥ": 163,
+  "Ĥ</w>": 262,
+  "ĥ": 164,
+  "ĥ</w>": 337,
+  "Ħ": 165,
+  "Ħ</w>": 278,
+  "ħ": 166,
+  "ħ</w>": 281,
+  "Ĩ": 167,
+  "Ĩ</w>": 308,
+  "ĩ": 168,
+  "ĩ</w>": 225,
+  "Ī": 169,
+  "Ī</w>": 221,
+  "ī": 170,
+  "ī</w>": 244,
+  "Ĭ": 171,
+  "Ĭ</w>": 315,
+  "ĭ": 172,
+  "ĭ</w>": 321,
+  "Į": 173,
+  "Į</w>": 324,
+  "į": 174,
+  "į</w>": 302,
+  "İ": 175,
+  "İ</w>": 249,
+  "ı": 176,
+  "ı</w>": 332,
+  "IJ": 177,
+  "IJ</w>": 295,
+  "ij": 178,
+  "ij</w>": 313,
+  "Ĵ": 179,
+  "Ĵ</w>": 328,
+  "ĵ": 180,
+  "ĵ</w>": 312,
+  "Ķ": 181,
+  "Ķ</w>": 256,
+  "ķ": 182,
+  "ķ</w>": 314,
+  "ĸ": 183,
+  "ĸ</w>": 277,
+  "Ĺ": 184,
+  "Ĺ</w>": 322,
+  "ĺ": 185,
+  "ĺ</w>": 285,
+  "Ļ": 186,
+  "Ļ</w>": 267,
+  "ļ": 187,
+  "ļ</w>": 290,
+  "Ľ": 188,
+  "Ľ</w>": 311,
+  "ľ": 189,
+  "ľ</w>": 299,
+  "Ŀ": 190,
+  "Ŀ</w>": 291,
+  "ŀ": 191,
+  "ŀ</w>": 293,
+  "Ł": 192,
+  "Ł</w>": 335,
+  "ł": 193,
+  "ł</w>": 252,
+  "Ń": 194,
+  "Ń</w>": 297
+}

tokenizer_2/merges.txt

@@ -0,0 +1,647 @@
+#version: 0.2
+Ġ t
+Ġt h
+Ġ a
+Ġth e</w>
+i n
+Ġ o
+Ġ ,</w>
+Ġ s
+e d</w>
+Ġ w
+e r
+Ġ .</w>
+Ġ i
+r e
+Ġ c
+n d</w>
+Ġ f
+Ġ b
+a t
+Ġo f</w>
+e r</w>
+e n
+a r
+o r
+i t
+Ġ p
+Ġ h
+Ġa nd</w>
+o n
+in g</w>
+a n
+r o
+Ġ m
+Ġ d
+e s</w>
+Ġi n</w>
+o n</w>
+Ġt o</w>
+o u
+i s
+Ġ a</w>
+i c
+Ġ T
+a l
+Ġ l
+Ġ =</w>
+Ġ re
+Ġ "</w>
+e s
+Ġ S
+a s</w>
+a l</w>
+i l
+e l
+i on</w>
+Ġ A
+Ġ C
+Ġ 1
+Ġ Ċ</w>
+u r
+ĠT h
+Ġ n
+a s
+Ġ @
+e c
+o m
+a c
+Ġ e
+Ġw as</w>
+Ġ M
+o r</w>
+a n</w>
+a m
+e n</w>
+o l
+Ġ in
+Ġ g
+Ġ '</w>
+Ġ B
+l y</w>
+a t</w>
+i v
+t s</w>
+ĠTh e</w>
+u s
+- @</w>
+Ġ@ -@</w>
+i s</w>
+Ġ I
+Ġw h
+i g
+Ġ H
+Ġs t
+o s
+u n
+t h
+Ġ P
+Ġw it
+Ġth at</w>
+i r
+Ġa s</w>
+e m
+Ġo n</w>
+r a
+Ġf or</w>
+Ġ R
+e t
+o w
+Ġ 2
+i d
+Ġ D
+l e</w>
+Ġwit h</w>
+l a
+en t</w>
+i m
+Ġ F
+e a
+i on
+Ġb y</w>
+Ġ )</w>
+Ġ (</w>
+Ġa l
+Ġc on
+en t
+Ġ W
+Ġi s</w>
+er e</w>
+Ġ G
+Ġ N
+Ġ L
+Ġh a
+er s</w>
+r i
+t h</w>
+t ed</w>
+u c
+Ġ J
+Ġ1 9
+e v
+u l
+Ġ v
+c e</w>
+at ion</w>
+ro m</w>
+Ġb e
+Ġ E
+i n</w>
+Ġth e
+Ġf rom</w>
+Ġ O
+t er</w>
+Ġp ro
+Ġa r
+a d
+Ġc om
+i c</w>
+a g
+Ġh is</w>
+Ġs h
+Ġa t</w>
+o v
+i es</w>
+o o
+p p
+s t
+c h
+Ġ r
+Ġ2 0
+a y</w>
+i f
+Ġw ere</w>
+Ġc h
+u t</w>
+s t</w>
+u t
+d s</w>
+o p
+u m
+Ġi t</w>
+o c
+t er
+l e
+ig h
+u d
+Ġe x
+ion s</w>
+at e</w>
+it y</w>
+at ed</w>
+Ġ un
+e p
+q u
+Ġn o
+Ġ K
+iv e</w>
+is t
+Ġo n
+am e</w>
+ou n
+i r</w>
+a b
+Ġ â
+in g
+Ġh e</w>
+l d</w>
+u g
+ic h</w>
+Ġa n</w>
+e d
+Ġ k
+Ġâ Ģ
+Ġha d</w>
+v e</w>
+a in
+Ġs e
+t ion</w>
+or e</w>
+re s
+Ġwh ich</w>
+ĠI n</w>
+o d
+th er</w>
+a k
+Ġs p
+a r</w>
+Ġ y
+ĠC h
+on g</w>
+Ġa c
+es t</w>
+Ġ U
+a p
+f f
+al ly</w>
+r it
+ĠS t
+u b
+g e</w>
+b er</w>
+e t</w>
+Ġb e</w>
+e ar
+Ġre c
+er s
+Ġf ir
+o t
+Ġar e</w>
+Ġa n
+c h</w>
+o g
+i a</w>
+es t
+in e</w>
+il l
+an d
+e l</w>
+ar y</w>
+e w</w>
+i d</w>
+Ġf or
+Ġ ;</w>
+Ġcom p
+Ġ V
+Ġin c
+t r
+Ġ20 0
+Ġthe ir</w>
+u s</w>
+Ġb ut</w>
+r an
+ic al</w>
+Ġfir st</w>
+Ġd e
+Ġin t
+Ġ ro
+s o</w>
+ĠâĢ ĵ</w>
+Ġno t</w>
+d ing</w>
+f ter</w>
+ur e</w>
+Ġp ar
+Ġ :</w>
+i an</w>
+Ġt w
+ou ld</w>
+Ġal so</w>
+Ġi ts</w>
+Ġw or
+u m</w>
+Ġo r</w>
+os t</w>
+0 0</w>
+ou r
+ar d</w>
+Ġre s
+m p
+u e</w>
+Ġa b
+is h</w>
+Ġcon t
+Ġa d
+ow n</w>
+al l</w>
+ou g
+Ġh er</w>
+as t</w>
+Ġ en
+om e</w>
+al l
+d ed</w>
+o w</w>
+Ġha ve</w>
+Ġ us
+ea r</w>
+ac k</w>
+d uc
+i al</w>
+s s
+en ts</w>
+a in</w>
+t ing</w>
+Ġon e</w>
+es s
+Ġh as</w>
+igh t</w>
+a v
+Ġe v
+ou t</w>
+a y
+en ce</w>
+Ġbe en</w>
+e w
+Ġtw o</w>
+Ġc l
+d er</w>
+im e</w>
+k s</w>
+es s</w>
+is h
+. @</w>
+Ġ@ .@</w>
+Ġp la
+Ġp l
+Ġo r
+u p</w>
+m ent</w>
+ur ing</w>
+ol l
+ĠI n
+Ġth is</w>
+Ġb ec
+Ġcom m
+Ġd is
+at er</w>
+ag e</w>
+Ġa pp
+ou s</w>
+e y</w>
+i l</w>
+p er
+ĠA l
+ion al</w>
+l ud
+el y</w>
+t t
+il e</w>
+i z
+Ġ j
+Ġwh o</w>
+Ġa g
+i b
+Ġthe y</w>
+f or
+Ġo v
+at h
+e g
+Ġs c
+i p
+Ġ20 1
+Ġ 3
+Ġp er
+or y</w>
+Ġd es
+id e</w>
+Ġs er
+s e</w>
+ĠH e</w>
+la nd</w>
+at ions</w>
+r ic
+i t</w>
+re s</w>
+er ed</w>
+Ġp re
+ĠS h
+an ce</w>
+or t</w>
+an t</w>
+, @</w>
+Ġ@ ,@</w>
+el l</w>
+Ġ Y
+n ed</w>
+el l
+it e</w>
+Ġinc lud
+Ġre p
+Ġa fter</w>
+Ġs uc
+re e</w>
+an y</w>
+i m</w>
+or t
+Ġ1 8
+Ġs u
+ad e</w>
+ou r</w>
+ĠU n
+ĠI t</w>
+i k
+ĠM ar
+em ber</w>
+Ġ 1</w>
+e en</w>
+a nd</w>
+Ġs ec
+ic e</w>
+Ġt ime</w>
+ĠA n
+Ġint o</w>
+Ġf in
+Ġo ther</w>
+Ġa tt
+il l</w>
+re n
+ac h
+as s
+er al</w>
+es e</w>
+s h
+al s</w>
+it ion</w>
+oug h</w>
+l es</w>
+am p
+Ġw ould</w>
+Ġm ore</w>
+ro ug
+ri b
+er y</w>
+ac e</w>
+Ġ A</w>
+Ġpla y
+it ed</w>
+k ed</w>
+is t</w>
+i ed</w>
+Ġ 2</w>
+as ed</w>
+ing s</w>
+an g
+a m</w>
+i p</w>
+Ġb o
+ab le</w>
+t y</w>
+Ġch ar
+Ġc ent
+et w
+at es</w>
+ro p
+Ġ I</w>
+u nd</w>
+ĠA m
+c es</w>
+o in
+Ġin ter
+u p
+c t
+on e</w>
+Ġt ra
+an t
+ec t
+Ġal l</w>
+e f
+Ġcon s
+ub l
+n ing</w>
+an s</w>
+Ġf e
+us t</w>
+Ġ 0
+Ġre m
+as e</w>
+on g
+Ġwh en</w>
+e b
+ĠW h
+Ġe ar
+ev er</w>
+Ġov er</w>
+Ġk n
+a us
+Ġp os
+a d</w>
+er m
+Ġsh e</w>
+Ġ ra
+Ġd uring</w>
+as on</w>
+v i
+Ġex p
+Ġl ea
+Ġ el
+Ġ 4
+Ġon ly</w>
+o nd</w>
+Ġd ec
+Ġac c
+Ġo ff
+is s
+Ġf l
+ĠE n
+o t</w>
+en s
+os e</w>
+ak e</w>
+o m</w>
+Ġs ev
+ac h</w>
+etw een</w>
+er n
+Ġ 3</w>
+Ġp r
+Ġg ro
+r uc
+Ġd i
+Ġ19 9
+ĠA r
+Ġg ame</w>
+Ġh im</w>
+oo k</w>
+Ġ up</w>
+Ġab out</w>
+Ġre l
+for m
+Ġth ree</w>
+at t
+ĠC om
+Ġs a
+ear s</w>
+Ġ 5
+r y</w>
+Ġi mp
+Ġm ost</w>
+f er
+Ġp res
+Ġf il
+Ġb etween</w>
+Ġbe g
+p h
+or s</w>
+Ġth an</w>
+Ġrec or
+o b
+er ic
+at ing</w>
+Ġth roug
+k ing</w>
+Ġo ut</w>
+Ġn um
+oo d</w>
+oll ow
+ac t
+u il
+Ġc re
+ol og
+at ional</w>
+Ġpro duc
+Ġwh ile</w>
+Ġl ater</w>
+Ġw rit
+e x
+Ġst ar
+Ġsp ec
+e e
+ish ed</w>
+Ġre g
+is ion</w>
+ou th</w>
+Ġre le
+Ġa ss
+Ġse ason</w>
+Ġm ade</w>
+il y</w>
+r u
+o y
+t ur
+t e</w>
+Ġ qu
+Ġm ov
+ur y</w>
+ĠAm eric
+em ent</w>
+c c
+ou nd</w>
+Ġl ar
+Ġfor m
+ec t</w>
+Ġde f
+Ġm us
+ĠP ar
+Ġm e
+Ġs ub
+w ay</w>
+o p</w>
+o h
+el d</w>
+i e</w>
+em p
+am es</w>
+er n</w>
+Ġn or
+iv ed</w>
+ev el
+Ġsuc h</w>
+ar ds</w>
+Ġin d
+ik e</w>
+Ġg en
+er t
+Ġy ear</w>
+Ġus ed</w>
+Ġn ew</w>
+Ġ 5</w>
+Ġal b
+s p
+y p
+Ġwit h
+Ġwh ere</w>
+ic s</w>
+ĠTh is</w>
+Ġthe m</w>
+w n</w>

tokenizer_2/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_2/tokenizer_config.json

@@ -0,0 +1,30 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|startoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 77,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": "<|endoftext|>"
+}

tokenizer_2/vocab.json

@@ -0,0 +1,1002 @@
+{
+  "!": 2,
+  "!</w>": 345,
+  "\"": 3,
+  "\"</w>": 344,
+  "#": 4,
+  "#</w>": 325,
+  "$": 5,
+  "$</w>": 348,
+  "%": 6,
+  "%</w>": 351,
+  "&": 7,
+  "&</w>": 352,
+  "'": 8,
+  "'</w>": 296,
+  "(": 9,
+  "(</w>": 318,
+  ")": 10,
+  ")</w>": 330,
+  "*": 11,
+  "*</w>": 327,
+  "+": 12,
+  "+</w>": 341,
+  ",": 13,
+  ",</w>": 279,
+  ",@</w>": 754,
+  "-": 14,
+  "-</w>": 276,
+  "-@</w>": 439,
+  ".": 15,
+  ".</w>": 253,
+  ".@</w>": 695,
+  "/": 16,
+  "/</w>": 350,
+  "0": 17,
+  "00</w>": 647,
+  "0</w>": 216,
+  "1": 18,
+  "1</w>": 222,
+  "2": 19,
+  "2</w>": 231,
+  "3": 20,
+  "3</w>": 243,
+  "4": 21,
+  "4</w>": 233,
+  "5": 22,
+  "5</w>": 240,
+  "6": 23,
+  "6</w>": 226,
+  "7": 24,
+  "7</w>": 215,
+  "8": 25,
+  "8</w>": 236,
+  "9": 26,
+  "9</w>": 242,
+  ":": 27,
+  ":</w>": 353,
+  ";": 28,
+  ";</w>": 317,
+  "<": 29,
+  "<</w>": 340,
+  "<|endoftext|>": 1,
+  "<|startoftext|>": 0,
+  "=": 30,
+  "=</w>": 342,
+  ">": 31,
+  "></w>": 300,
+  "?": 32,
+  "?</w>": 346,
+  "@": 33,
+  "@</w>": 320,
+  "A": 34,
+  "A</w>": 227,
+  "B": 35,
+  "B</w>": 258,
+  "C": 36,
+  "C</w>": 239,
+  "D": 37,
+  "D</w>": 255,
+  "E": 38,
+  "E</w>": 246,
+  "F": 39,
+  "F</w>": 213,
+  "G": 40,
+  "G</w>": 283,
+  "H": 41,
+  "H</w>": 219,
+  "I": 42,
+  "I</w>": 237,
+  "J": 43,
+  "J</w>": 251,
+  "K": 44,
+  "K</w>": 254,
+  "L": 45,
+  "L</w>": 218,
+  "M": 46,
+  "M</w>": 234,
+  "N": 47,
+  "N</w>": 238,
+  "O": 48,
+  "O</w>": 265,
+  "P": 49,
+  "P</w>": 245,
+  "Q": 50,
+  "Q</w>": 309,
+  "R": 51,
+  "R</w>": 264,
+  "S": 52,
+  "S</w>": 230,
+  "T": 53,
+  "T</w>": 235,
+  "U": 54,
+  "U</w>": 268,
+  "V": 55,
+  "V</w>": 248,
+  "W": 56,
+  "W</w>": 274,
+  "X": 57,
+  "X</w>": 263,
+  "Y": 58,
+  "Y</w>": 310,
+  "Z": 59,
+  "Z</w>": 207,
+  "[": 60,
+  "[</w>": 270,
+  "\\": 61,
+  "\\</w>": 338,
+  "]": 62,
+  "]</w>": 289,
+  "^": 63,
+  "^</w>": 331,
+  "_": 64,
+  "_</w>": 334,
+  "`": 65,
+  "`</w>": 347,
+  "a": 66,
+  "a</w>": 197,
+  "ab": 555,
+  "able</w>": 820,
+  "ac": 420,
+  "ace</w>": 806,
+  "ach": 791,
+  "ach</w>": 885,
+  "ack</w>": 670,
+  "act": 929,
+  "ad": 508,
+  "ad</w>": 860,
+  "ade</w>": 771,
+  "ag": 511,
+  "age</w>": 710,
+  "ain": 568,
+  "ain</w>": 675,
+  "ak": 577,
+  "ake</w>": 882,
+  "al": 397,
+  "al</w>": 405,
+  "all": 664,
+  "all</w>": 658,
+  "ally</w>": 588,
+  "als</w>": 796,
+  "am": 426,
+  "am</w>": 817,
+  "ame</w>": 552,
+  "ames</w>": 976,
+  "amp": 800,
+  "an": 384,
+  "an</w>": 425,
+  "ance</w>": 751,
+  "and": 609,
+  "and</w>": 780,
+  "ang": 816,
+  "ans</w>": 844,
+  "ant": 837,
+  "ant</w>": 753,
+  "any</w>": 766,
+  "ap": 586,
+  "ar": 376,
+  "ar</w>": 579,
+  "ard</w>": 649,
+  "ards</w>": 982,
+  "ary</w>": 611,
+  "as": 416,
+  "as</w>": 404,
+  "ase</w>": 849,
+  "ased</w>": 814,
+  "ason</w>": 865,
+  "ass": 792,
+  "ast</w>": 661,
+  "at": 372,
+  "at</w>": 434,
+  "ate</w>": 541,
+  "ated</w>": 543,
+  "ater</w>": 709,
+  "ates</w>": 825,
+  "ath": 730,
+  "ating</w>": 922,
+  "ation</w>": 497,
+  "ational</w>": 933,
+  "ations</w>": 744,
+  "att": 903,
+  "aus": 858,
+  "av": 681,
+  "ay": 684,
+  "ay</w>": 523,
+  "b": 67,
+  "b</w>": 212,
+  "ber</w>": 593,
+  "c": 68,
+  "c</w>": 224,
+  "cc": 960,
+  "ce</w>": 496,
+  "ces</w>": 830,
+  "ch": 520,
+  "ch</w>": 603,
+  "ct": 834,
+  "d": 69,
+  "d</w>": 196,
+  "ded</w>": 665,
+  "der</w>": 690,
+  "ding</w>": 633,
+  "ds</w>": 530,
+  "duc": 671,
+  "e": 70,
+  "e</w>": 195,
+  "ea": 471,
+  "ear": 596,
+  "ear</w>": 669,
+  "ears</w>": 906,
+  "eb": 852,
+  "ec": 418,
+  "ect": 838,
+  "ect</w>": 964,
+  "ed": 563,
+  "ed</w>": 362,
+  "ee": 941,
+  "een</w>": 779,
+  "ef": 840,
+  "eg": 731,
+  "el": 407,
+  "el</w>": 610,
+  "eld</w>": 973,
+  "ell": 759,
+  "ell</w>": 756,
+  "ely</w>": 719,
+  "em": 455,
+  "ember</w>": 777,
+  "ement</w>": 959,
+  "emp": 975,
+  "en": 375,
+  "en</w>": 427,
+  "ence</w>": 685,
+  "ens": 880,
+  "ent": 478,
+  "ent</w>": 468,
+  "ents</w>": 674,
+  "ep": 545,
+  "er": 364,
+  "er</w>": 374,
+  "eral</w>": 793,
+  "ere</w>": 481,
+  "ered</w>": 748,
+  "eric": 921,
+  "erm": 861,
+  "ern": 887,
+  "ern</w>": 977,
+  "ers": 598,
+  "ers</w>": 486,
+  "ert": 986,
+  "ery</w>": 805,
+  "es": 402,
+  "es</w>": 388,
+  "ese</w>": 794,
+  "ess": 678,
+  "ess</w>": 693,
+  "est": 606,
+  "est</w>": 584,
+  "et": 460,
+  "et</w>": 594,
+  "etw": 824,
+  "etween</w>": 886,
+  "ev": 493,
+  "evel": 980,
+  "ever</w>": 855,
+  "ew": 687,
+  "ew</w>": 612,
+  "ex": 938,
+  "ey</w>": 713,
+  "f": 71,
+  "f</w>": 209,
+  "fer": 911,
+  "ff": 587,
+  "for": 728,
+  "form": 901,
+  "fter</w>": 634,
+  "g": 72,
+  "g</w>": 214,
+  "ge</w>": 592,
+  "h": 73,
+  "h</w>": 203,
+  "i": 74,
+  "i</w>": 205,
+  "ia</w>": 605,
+  "ial</w>": 672,
+  "ian</w>": 638,
+  "ib": 726,
+  "ic": 395,
+  "ic</w>": 510,
+  "ical</w>": 625,
+  "ice</w>": 782,
+  "ich</w>": 561,
+  "ics</w>": 996,
+  "id": 463,
+  "id</w>": 613,
+  "ide</w>": 739,
+  "ie</w>": 974,
+  "ied</w>": 812,
+  "ies</w>": 516,
+  "if": 524,
+  "ig": 444,
+  "igh": 537,
+  "ight</w>": 680,
+  "ik": 775,
+  "ike</w>": 984,
+  "il": 406,
+  "il</w>": 714,
+  "ile</w>": 721,
+  "ill": 608,
+  "ill</w>": 789,
+  "ily</w>": 950,
+  "im": 469,
+  "im</w>": 767,
+  "ime</w>": 691,
+  "in": 358,
+  "in</w>": 501,
+  "ine</w>": 607,
+  "ing": 557,
+  "ing</w>": 383,
+  "ings</w>": 815,
+  "ion": 472,
+  "ion</w>": 408,
+  "ional</w>": 717,
+  "ions</w>": 540,
+  "ip": 733,
+  "ip</w>": 818,
+  "ir": 453,
+  "ir</w>": 554,
+  "is": 393,
+  "is</w>": 441,
+  "ish": 694,
+  "ish</w>": 654,
+  "ished</w>": 942,
+  "ision</w>": 944,
+  "iss": 876,
+  "ist": 550,
+  "ist</w>": 811,
+  "it": 378,
+  "it</w>": 746,
+  "ite</w>": 760,
+  "ited</w>": 809,
+  "ition</w>": 797,
+  "ity</w>": 542,
+  "iv": 435,
+  "ive</w>": 549,
+  "ived</w>": 979,
+  "iz": 722,
+  "j": 75,
+  "j</w>": 288,
+  "k": 76,
+  "k</w>": 210,
+  "ked</w>": 810,
+  "king</w>": 924,
+  "ks</w>": 692,
+  "l": 77,
+  "l</w>": 201,
+  "la": 467,
+  "land</w>": 743,
+  "ld</w>": 559,
+  "le": 536,
+  "le</w>": 465,
+  "les</w>": 799,
+  "lud": 718,
+  "ly</w>": 433,
+  "m": 78,
+  "m</w>": 202,
+  "ment</w>": 701,
+  "mp": 651,
+  "n": 79,
+  "n</w>": 199,
+  "nd</w>": 369,
+  "ned</w>": 758,
+  "ning</w>": 843,
+  "o": 80,
+  "o</w>": 198,
+  "ob": 920,
+  "oc": 534,
+  "od": 575,
+  "og": 604,
+  "oh": 972,
+  "oin": 831,
+  "ol": 428,
+  "oll": 703,
+  "ollow": 928,
+  "olog": 932,
+  "om": 419,
+  "om</w>": 883,
+  "ome</w>": 663,
+  "on": 382,
+  "on</w>": 390,
+  "ond</w>": 872,
+  "one</w>": 835,
+  "ong": 850,
+  "ong</w>": 582,
+  "oo": 517,
+  "ood</w>": 927,
+  "ook</w>": 897,
+  "op": 531,
+  "op</w>": 971,
+  "or": 377,
+  "or</w>": 424,
+  "ore</w>": 571,
+  "ors</w>": 917,
+  "ort": 768,
+  "ort</w>": 752,
+  "ory</w>": 737,
+  "os": 447,
+  "ose</w>": 881,
+  "ost</w>": 646,
+  "ot": 600,
+  "ot</w>": 879,
+  "ou": 392,
+  "oug": 659,
+  "ough</w>": 798,
+  "ould</w>": 640,
+  "oun": 553,
+  "ound</w>": 961,
+  "our": 648,
+  "our</w>": 772,
+  "ous</w>": 712,
+  "out</w>": 683,
+  "outh</w>": 945,
+  "ov": 515,
+  "ow": 461,
+  "ow</w>": 666,
+  "own</w>": 657,
+  "oy": 952,
+  "p": 81,
+  "p</w>": 217,
+  "per": 715,
+  "ph": 916,
+  "pp": 518,
+  "q": 82,
+  "q</w>": 280,
+  "qu": 546,
+  "r": 83,
+  "r</w>": 204,
+  "ra": 457,
+  "ran": 624,
+  "re": 367,
+  "ree</w>": 765,
+  "ren": 790,
+  "res": 572,
+  "res</w>": 747,
+  "ri": 487,
+  "rib": 804,
+  "ric": 745,
+  "rit": 589,
+  "ro": 385,
+  "rom</w>": 498,
+  "rop": 826,
+  "roug": 803,
+  "ru": 951,
+  "ruc": 891,
+  "ry</w>": 908,
+  "s": 84,
+  "s</w>": 206,
+  "se</w>": 741,
+  "sh": 795,
+  "so</w>": 630,
+  "sp": 992,
+  "ss": 673,
+  "st": 519,
+  "st</w>": 528,
+  "t": 85,
+  "t</w>": 208,
+  "te</w>": 954,
+  "ted</w>": 489,
+  "ter": 535,
+  "ter</w>": 505,
+  "th": 449,
+  "th</w>": 488,
+  "ther</w>": 576,
+  "ting</w>": 676,
+  "tion</w>": 570,
+  "tr": 619,
+  "ts</w>": 436,
+  "tt": 720,
+  "tur": 953,
+  "ty</w>": 821,
+  "u": 86,
+  "u</w>": 229,
+  "ub": 591,
+  "ubl": 842,
+  "uc": 490,
+  "ud": 538,
+  "ue</w>": 652,
+  "ug": 560,
+  "uil": 930,
+  "ul": 494,
+  "um": 532,
+  "um</w>": 644,
+  "un": 448,
+  "und</w>": 828,
+  "up": 833,
+  "up</w>": 700,
+  "ur": 413,
+  "ure</w>": 635,
+  "uring</w>": 702,
+  "ury</w>": 957,
+  "us": 438,
+  "us</w>": 622,
+  "ust</w>": 846,
+  "ut": 529,
+  "ut</w>": 527,
+  "v": 87,
+  "v</w>": 232,
+  "ve</w>": 567,
+  "vi": 866,
+  "w": 88,
+  "w</w>": 250,
+  "way</w>": 970,
+  "wn</w>": 999,
+  "x": 89,
+  "x</w>": 269,
+  "y": 90,
+  "y</w>": 211,
+  "yp": 993,
+  "z": 91,
+  "z</w>": 228,
+  "|": 92,
+  "|</w>": 304,
+  "}": 93,
+  "}</w>": 336,
+  "~": 94,
+  "~</w>": 343,
+  "¡": 95,
+  "¡</w>": 220,
+  "¢": 96,
+  "¢</w>": 306,
+  "£": 97,
+  "£</w>": 323,
+  "¤": 98,
+  "¤</w>": 292,
+  "¥": 99,
+  "¥</w>": 339,
+  "¦": 100,
+  "¦</w>": 303,
+  "§": 101,
+  "§</w>": 275,
+  "¨": 102,
+  "¨</w>": 282,
+  "©": 103,
+  "©</w>": 259,
+  "ª": 104,
+  "ª</w>": 286,
+  "«": 105,
+  "«</w>": 266,
+  "¬": 106,
+  "¬</w>": 319,
+  "®": 107,
+  "®</w>": 329,
+  "¯": 108,
+  "¯</w>": 287,
+  "°": 109,
+  "°</w>": 298,
+  "±": 110,
+  "±</w>": 200,
+  "²": 111,
+  "²</w>": 284,
+  "³": 112,
+  "³</w>": 272,
+  "´": 113,
+  "´</w>": 307,
+  "µ": 114,
+  "µ</w>": 261,
+  "¶": 115,
+  "¶</w>": 301,
+  "·": 116,
+  "·</w>": 326,
+  "¸": 117,
+  "¸</w>": 257,
+  "¹": 118,
+  "¹</w>": 241,
+  "º": 119,
+  "º</w>": 260,
+  "»": 120,
+  "»</w>": 247,
+  "¼": 121,
+  "¼</w>": 305,
+  "½": 122,
+  "½</w>": 294,
+  "¾": 123,
+  "¾</w>": 316,
+  "¿": 124,
+  "¿</w>": 271,
+  "Â": 125,
+  "Ã": 126,
+  "Ä": 127,
+  "Å": 128,
+  "Æ": 129,
+  "Ç": 130,
+  "È": 131,
+  "É": 132,
+  "Ê": 133,
+  "Ë": 134,
+  "Ì": 135,
+  "Í": 136,
+  "Î": 137,
+  "Ï": 138,
+  "Ð": 139,
+  "Ñ": 140,
+  "Ö": 141,
+  "×": 142,
+  "Ø": 143,
+  "Ù": 144,
+  "Ü": 145,
+  "à": 146,
+  "á": 147,
+  "â": 148,
+  "ã": 149,
+  "ä": 150,
+  "å": 151,
+  "æ": 152,
+  "ç": 153,
+  "è": 154,
+  "é": 155,
+  "ë": 156,
+  "ì": 157,
+  "ï": 158,
+  "Ċ": 159,
+  "Ċ</w>": 349,
+  "Ġ": 160,
+  "Ġ\"</w>": 401,
+  "Ġ'</w>": 431,
+  "Ġ(</w>": 475,
+  "Ġ)</w>": 474,
+  "Ġ,</w>": 360,
+  "Ġ.</w>": 365,
+  "Ġ0": 847,
+  "Ġ1": 411,
+  "Ġ18": 769,
+  "Ġ19": 492,
+  "Ġ199": 893,
+  "Ġ1</w>": 778,
+  "Ġ2": 462,
+  "Ġ20": 522,
+  "Ġ200": 620,
+  "Ġ201": 734,
+  "Ġ2</w>": 813,
+  "Ġ3": 735,
+  "Ġ3</w>": 888,
+  "Ġ4": 870,
+  "Ġ5": 907,
+  "Ġ5</w>": 990,
+  "Ġ:</w>": 637,
+  "Ġ;</w>": 615,
+  "Ġ</w>": 333,
+  "Ġ=</w>": 399,
+  "Ġ@": 417,
+  "Ġ@,@</w>": 755,
+  "Ġ@-@</w>": 440,
+  "Ġ@.@</w>": 696,
+  "ĠA": 409,
+  "ĠA</w>": 807,
+  "ĠAl": 716,
+  "ĠAm": 829,
+  "ĠAmeric": 958,
+  "ĠAn": 784,
+  "ĠAr": 894,
+  "ĠB": 432,
+  "ĠC": 410,
+  "ĠCh": 581,
+  "ĠCom": 904,
+  "ĠD": 464,
+  "ĠE": 500,
+  "ĠEn": 878,
+  "ĠF": 470,
+  "ĠG": 482,
+  "ĠH": 445,
+  "ĠHe</w>": 742,
+  "ĠI": 442,
+  "ĠI</w>": 827,
+  "ĠIn": 704,
+  "ĠIn</w>": 574,
+  "ĠIt</w>": 774,
+  "ĠJ": 491,
+  "ĠK": 548,
+  "ĠL": 484,
+  "ĠM": 423,
+  "ĠMar": 776,
+  "ĠN": 483,
+  "ĠO": 504,
+  "ĠP": 450,
+  "ĠPar": 967,
+  "ĠR": 459,
+  "ĠS": 403,
+  "ĠSh": 750,
+  "ĠSt": 590,
+  "ĠT": 396,
+  "ĠTh": 414,
+  "ĠThe</w>": 437,
+  "ĠThis</w>": 997,
+  "ĠU": 585,
+  "ĠUn": 773,
+  "ĠV": 617,
+  "ĠW": 479,
+  "ĠWh": 853,
+  "ĠY": 757,
+  "Ġa": 356,
+  "Ġa</w>": 394,
+  "Ġab": 653,
+  "Ġabout</w>": 899,
+  "Ġac": 583,
+  "Ġacc": 874,
+  "Ġad": 656,
+  "Ġafter</w>": 763,
+  "Ġag": 725,
+  "Ġal": 476,
+  "Ġalb": 991,
+  "Ġall</w>": 839,
+  "Ġalso</w>": 641,
+  "Ġan": 602,
+  "Ġan</w>": 562,
+  "Ġand</w>": 381,
+  "Ġapp": 711,
+  "Ġar": 507,
+  "Ġare</w>": 601,
+  "Ġas</w>": 454,
+  "Ġass": 947,
+  "Ġat</w>": 514,
+  "Ġatt": 788,
+  "Ġb": 371,
+  "Ġbe": 499,
+  "Ġbe</w>": 595,
+  "Ġbec": 706,
+  "Ġbeen</w>": 686,
+  "Ġbeg": 915,
+  "Ġbetween</w>": 914,
+  "Ġbo": 819,
+  "Ġbut</w>": 623,
+  "Ġby</w>": 473,
+  "Ġc": 368,
+  "Ġcent": 823,
+  "Ġch": 526,
+  "Ġchar": 822,
+  "Ġcl": 689,
+  "Ġcom": 509,
+  "Ġcomm": 707,
+  "Ġcomp": 616,
+  "Ġcon": 477,
+  "Ġcons": 841,
+  "Ġcont": 655,
+  "Ġcre": 931,
+  "Ġd": 387,
+  "Ġde": 627,
+  "Ġdec": 873,
+  "Ġdef": 965,
+  "Ġdes": 738,
+  "Ġdi": 892,
+  "Ġdis": 708,
+  "Ġduring</w>": 864,
+  "Ġe": 421,
+  "Ġear": 854,
+  "Ġel": 869,
+  "Ġen": 662,
+  "Ġev": 682,
+  "Ġex": 539,
+  "Ġexp": 867,
+  "Ġf": 370,
+  "Ġfe": 845,
+  "Ġfil": 913,
+  "Ġfin": 786,
+  "Ġfir": 599,
+  "Ġfirst</w>": 626,
+  "Ġfl": 877,
+  "Ġfor": 614,
+  "Ġfor</w>": 458,
+  "Ġform": 963,
+  "Ġfrom</w>": 503,
+  "Ġg": 430,
+  "Ġgame</w>": 895,
+  "Ġgen": 985,
+  "Ġgro": 890,
+  "Ġh": 380,
+  "Ġha": 485,
+  "Ġhad</w>": 566,
+  "Ġhas</w>": 679,
+  "Ġhave</w>": 667,
+  "Ġhe</w>": 558,
+  "Ġher</w>": 660,
+  "Ġhim</w>": 896,
+  "Ġhis</w>": 512,
+  "Ġi": 366,
+  "Ġimp": 909,
+  "Ġin": 429,
+  "Ġin</w>": 389,
+  "Ġinc": 618,
+  "Ġinclud": 761,
+  "Ġind": 983,
+  "Ġint": 628,
+  "Ġinter": 832,
+  "Ġinto</w>": 785,
+  "Ġis</w>": 480,
+  "Ġit</w>": 533,
+  "Ġits</w>": 642,
+  "Ġj": 723,
+  "Ġk": 564,
+  "Ġkn": 857,
+  "Ġl": 398,
+  "Ġlar": 962,
+  "Ġlater</w>": 936,
+  "Ġlea": 868,
+  "Ġm": 386,
+  "Ġmade</w>": 949,
+  "Ġme": 968,
+  "Ġmore</w>": 802,
+  "Ġmost</w>": 910,
+  "Ġmov": 956,
+  "Ġmus": 966,
+  "Ġn": 415,
+  "Ġnew</w>": 989,
+  "Ġno": 547,
+  "Ġnor": 978,
+  "Ġnot</w>": 632,
+  "Ġnum": 926,
+  "Ġo": 359,
+  "Ġof</w>": 373,
+  "Ġoff": 875,
+  "Ġon": 551,
+  "Ġon</w>": 456,
+  "Ġone</w>": 677,
+  "Ġonly</w>": 871,
+  "Ġor": 699,
+  "Ġor</w>": 645,
+  "Ġother</w>": 787,
+  "Ġout</w>": 925,
+  "Ġov": 729,
+  "Ġover</w>": 856,
+  "Ġp": 379,
+  "Ġpar": 636,
+  "Ġper": 736,
+  "Ġpl": 698,
+  "Ġpla": 697,
+  "Ġplay": 808,
+  "Ġpos": 859,
+  "Ġpr": 889,
+  "Ġpre": 749,
+  "Ġpres": 912,
+  "Ġpro": 506,
+  "Ġproduc": 934,
+  "Ġqu": 955,
+  "Ġr": 521,
+  "Ġra": 863,
+  "Ġre": 400,
+  "Ġrec": 597,
+  "Ġrecor": 919,
+  "Ġreg": 943,
+  "Ġrel": 900,
+  "Ġrele": 946,
+  "Ġrem": 848,
+  "Ġrep": 762,
+  "Ġres": 650,
+  "Ġro": 629,
+  "Ġs": 361,
+  "Ġsa": 905,
+  "Ġsc": 732,
+  "Ġse": 569,
+  "Ġseason</w>": 948,
+  "Ġsec": 781,
+  "Ġser": 740,
+  "Ġsev": 884,
+  "Ġsh": 513,
+  "Ġshe</w>": 862,
+  "Ġsp": 578,
+  "Ġspec": 940,
+  "Ġst": 446,
+  "Ġstar": 939,
+  "Ġsu": 770,
+  "Ġsub": 969,
+  "Ġsuc": 764,
+  "Ġsuch</w>": 981,
+  "Ġt": 354,
+  "Ġth": 355,
+  "Ġthan</w>": 918,
+  "Ġthat</w>": 452,
+  "Ġthe": 502,
+  "Ġthe</w>": 357,
+  "Ġtheir</w>": 621,
+  "Ġthem</w>": 998,
+  "Ġthey</w>": 727,
+  "Ġthis</w>": 705,
+  "Ġthree</w>": 902,
+  "Ġthroug": 923,
+  "Ġtime</w>": 783,
+  "Ġto</w>": 391,
+  "Ġtra": 836,
+  "Ġtw": 639,
+  "Ġtwo</w>": 688,
+  "Ġun": 544,
+  "Ġup</w>": 898,
+  "Ġus": 668,
+  "Ġused</w>": 988,
+  "Ġv": 495,
+  "Ġw": 363,
+  "Ġwas</w>": 422,
+  "Ġwere</w>": 525,
+  "Ġwh": 443,
+  "Ġwhen</w>": 851,
+  "Ġwhere</w>": 995,
+  "Ġwhich</w>": 573,
+  "Ġwhile</w>": 935,
+  "Ġwho</w>": 724,
+  "Ġwit": 451,
+  "Ġwith": 994,
+  "Ġwith</w>": 466,
+  "Ġwor": 643,
+  "Ġwould</w>": 801,
+  "Ġwrit": 937,
+  "Ġy": 580,
+  "Ġyear</w>": 987,
+  "Ġâ": 556,
+  "ĠâĢ": 565,
+  "ĠâĢĵ</w>": 631,
+  "ĠĊ</w>": 412,
+  "Ģ": 161,
+  "Ģ</w>": 223,
+  "ģ": 162,
+  "ģ</w>": 273,
+  "Ĥ": 163,
+  "Ĥ</w>": 262,
+  "ĥ": 164,
+  "ĥ</w>": 337,
+  "Ħ": 165,
+  "Ħ</w>": 278,
+  "ħ": 166,
+  "ħ</w>": 281,
+  "Ĩ": 167,
+  "Ĩ</w>": 308,
+  "ĩ": 168,
+  "ĩ</w>": 225,
+  "Ī": 169,
+  "Ī</w>": 221,
+  "ī": 170,
+  "ī</w>": 244,
+  "Ĭ": 171,
+  "Ĭ</w>": 315,
+  "ĭ": 172,
+  "ĭ</w>": 321,
+  "Į": 173,
+  "Į</w>": 324,
+  "į": 174,
+  "į</w>": 302,
+  "İ": 175,
+  "İ</w>": 249,
+  "ı": 176,
+  "ı</w>": 332,
+  "IJ": 177,
+  "IJ</w>": 295,
+  "ij": 178,
+  "ij</w>": 313,
+  "Ĵ": 179,
+  "Ĵ</w>": 328,
+  "ĵ": 180,
+  "ĵ</w>": 312,
+  "Ķ": 181,
+  "Ķ</w>": 256,
+  "ķ": 182,
+  "ķ</w>": 314,
+  "ĸ": 183,
+  "ĸ</w>": 277,
+  "Ĺ": 184,
+  "Ĺ</w>": 322,
+  "ĺ": 185,
+  "ĺ</w>": 285,
+  "Ļ": 186,
+  "Ļ</w>": 267,
+  "ļ": 187,
+  "ļ</w>": 290,
+  "Ľ": 188,
+  "Ľ</w>": 311,
+  "ľ": 189,
+  "ľ</w>": 299,
+  "Ŀ": 190,
+  "Ŀ</w>": 291,
+  "ŀ": 191,
+  "ŀ</w>": 293,
+  "Ł": 192,
+  "Ł</w>": 335,
+  "ł": 193,
+  "ł</w>": 252,
+  "Ń": 194,
+  "Ń</w>": 297
+}

tokenizer_3/special_tokens_map.json

@@ -0,0 +1,125 @@
+{
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_3/tokenizer_config.json

@@ -0,0 +1,937 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<extra_id_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<extra_id_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<extra_id_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<extra_id_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<extra_id_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<extra_id_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<extra_id_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<extra_id_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<extra_id_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<extra_id_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<extra_id_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<extra_id_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<extra_id_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<extra_id_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<extra_id_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<extra_id_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<extra_id_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<extra_id_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<extra_id_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<extra_id_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<extra_id_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<extra_id_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<extra_id_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<extra_id_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<extra_id_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<extra_id_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<extra_id_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<extra_id_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<extra_id_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<extra_id_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<extra_id_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<extra_id_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<extra_id_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<extra_id_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<extra_id_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "38": {
+      "content": "<extra_id_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39": {
+      "content": "<extra_id_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40": {
+      "content": "<extra_id_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "41": {
+      "content": "<extra_id_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "42": {
+      "content": "<extra_id_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "43": {
+      "content": "<extra_id_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "44": {
+      "content": "<extra_id_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "45": {
+      "content": "<extra_id_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "46": {
+      "content": "<extra_id_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "47": {
+      "content": "<extra_id_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "48": {
+      "content": "<extra_id_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49": {
+      "content": "<extra_id_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50": {
+      "content": "<extra_id_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "51": {
+      "content": "<extra_id_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "52": {
+      "content": "<extra_id_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "53": {
+      "content": "<extra_id_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "54": {
+      "content": "<extra_id_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "55": {
+      "content": "<extra_id_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "56": {
+      "content": "<extra_id_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "57": {
+      "content": "<extra_id_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "58": {
+      "content": "<extra_id_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59": {
+      "content": "<extra_id_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "60": {
+      "content": "<extra_id_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61": {
+      "content": "<extra_id_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "62": {
+      "content": "<extra_id_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "63": {
+      "content": "<extra_id_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64": {
+      "content": "<extra_id_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "65": {
+      "content": "<extra_id_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "66": {
+      "content": "<extra_id_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "67": {
+      "content": "<extra_id_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "68": {
+      "content": "<extra_id_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "69": {
+      "content": "<extra_id_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "70": {
+      "content": "<extra_id_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "71": {
+      "content": "<extra_id_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "72": {
+      "content": "<extra_id_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "73": {
+      "content": "<extra_id_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "74": {
+      "content": "<extra_id_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "75": {
+      "content": "<extra_id_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "76": {
+      "content": "<extra_id_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "77": {
+      "content": "<extra_id_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "78": {
+      "content": "<extra_id_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<extra_id_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<extra_id_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "81": {
+      "content": "<extra_id_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "82": {
+      "content": "<extra_id_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "83": {
+      "content": "<extra_id_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "84": {
+      "content": "<extra_id_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "85": {
+      "content": "<extra_id_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86": {
+      "content": "<extra_id_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "87": {
+      "content": "<extra_id_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "88": {
+      "content": "<extra_id_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "89": {
+      "content": "<extra_id_13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "90": {
+      "content": "<extra_id_12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "91": {
+      "content": "<extra_id_11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92": {
+      "content": "<extra_id_10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "93": {
+      "content": "<extra_id_9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "94": {
+      "content": "<extra_id_8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "95": {
+      "content": "<extra_id_7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "96": {
+      "content": "<extra_id_6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "97": {
+      "content": "<extra_id_5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "98": {
+      "content": "<extra_id_4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "99": {
+      "content": "<extra_id_3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "<extra_id_2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "<extra_id_1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<extra_id_0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_ids": 100,
+  "model_max_length": 1024,
+  "pad_token": "<pad>",
+  "tokenizer_class": "T5Tokenizer",
+  "unk_token": "<unk>"
+}

transformer/config.json

@@ -0,0 +1,20 @@
+{
+  "_class_name": "SD3Transformer2DModel",
+  "_diffusers_version": "0.32.0.dev0",
+  "attention_head_dim": 8,
+  "caption_projection_dim": 32,
+  "dual_attention_layers": [
+    0,
+    1
+  ],
+  "in_channels": 8,
+  "joint_attention_dim": 32,
+  "num_attention_heads": 4,
+  "num_layers": 4,
+  "out_channels": 8,
+  "patch_size": 1,
+  "pooled_projection_dim": 64,
+  "pos_embed_max_size": 96,
+  "qk_norm": "rms_norm",
+  "sample_size": 32
+}

transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9ac2dc85c0926f3fe470aa8fa49cddbc961cbb7d6916a8b00da53b9b85e39fd
+size 1871464

vae/config.json

@@ -0,0 +1,28 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.32.0.dev0",
+  "act_fn": "silu",
+  "block_out_channels": [
+    4
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 8,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 1,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 1,
+  "out_channels": 3,
+  "sample_size": 32,
+  "scaling_factor": 1.5035,
+  "shift_factor": 0.0609,
+  "up_block_types": [
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": false,
+  "use_quant_conv": false
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1540138f8b1fb8927b746d7fdd94204ad18a8343da29e26b212c417c1a718351
+size 22652