Upload folder using huggingface_hub

.gitignore

@@ -0,0 +1,4 @@
+venv/
+__pycache__/
+*.pyc
+.env

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Tarush Gupta
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,3 +1,246 @@
----
-license: mit
----
+# AureliusGPT-Torch
+
+## Overview
+
+AureliusGPT-Torch is an 845k, PyTorch and SentencePiece boosted SLM pretrained on _Meditations_ by Marcus Aurelius and other adjacent philosophical works. It is a larger size, more comprehensive reimplementation of 
+[AureliusGPT](https://github.com/tarush-ai/AureliusGPT), a smaller model trained on the same core corpus, but using a handwritten/NumPy first (zero PyTorch or prewritten tokenizer) approach.
+
+Rather than reimplementing a custom BPE algorithm and tokenizer backpropagation from scratch like its sibling repo, AureliusGPT-Torch trains SentencePiece on its corpus (detailed in Data) and relies on PyTorch for autograd.
+
+The [HuggingFace](https://huggingface.co/Tarush-AI/AureliusGPT) for this model (including both the base model weights at epoch 10 and the tokenizer/vocabulary) are contained here. 
+
+## Usage
+
+To get started, run:
+
+```
+python -m venv venv
+source venv/bin/activate
+pip install -r requirements.txt
+```
+
+If you wish to have a "Stoic Validator," please create a .env file of the format:
+```
+OPENAI_API_KEY = sk_proj_YOUR_OPENAI_API_KEY_HERE
+```
+
+Alternatively, you can run from CLI and make the necessary changes:
+```
+export OPENAI_API_KEY=sk_proj_YOUR_OPENAI_API_KEY_HERE
+```
+
+A future version of this model will rely on the finetuned Llama 3.2 1B parameter model as the validator. 
+
+To finetune the teacher model for distillation / synthetic data generation, open the Colab Notebook (model/datageneration/llama321b_meditations_lora_unsloth.ipynb) and run it on the provided Tesla T4 GPU. If the data has not been generated, training will rely on the core _Meditations_ corpus. Future versions of this repository will have the synthetic data generation pipeline / Unsloth package library wrapped into the core repository with GPU plugin/handling.
+
+To run training from the source directory, run:
+```
+python -m model.train
+```
+
+To run the model in CLI, run:
+```
+python -m model.run
+```
+
+If you want to experiment with the preprocessing or tokenization logic in the corpus, .test() functions have been wrapped into both. You can try both:
+
+```
+python -m model.vocabulary.tokenizer test
+python -m model.vocabulary.preprocess test 
+```
+
+Or, if you have your own unique testfile (this may yield errors due to the hyperspecific nature of these files:
+```
+python -m model.vocabulary.tokenizer test YOUR_RELATIVE_FILEPATH_HERE
+python -m model.vocabulary.preprocess test YOUR_RELATIVE_FILEPATH_HERE
+```
+## Contribution
+
+This project uses an MIT license. You are more than welcome to submit a pull request and contribute on this repository. My future plans for this repository are listed under the "Future Work on AureliusGPT" section, but as long as the pull request is conducive, I will accept it. You can also reach out to me [here](mailto:tarushgs@gmail.com) to join as a contributor to the repository.
+
+## Data
+The original corpus of _Meditations_ by Marcus Aurelius is 89k tokens approximately when tokenized by a SentencePiece BPE tokenizer trained on a vocabulary length of 2,000. Using Kaplan et al. Chinchilla scaling laws,
+the expected parameter size of the model would be 8.9k parameters (taking the less conservative 1:10 ratio of parameters to corpus tokens). However, given the smaller size of the model and its lack of focus on general intelligence (instead, focused on generating Stoic-adjacent, Aurelius flavored text), this ratio does not apply.
+
+Given the risk of these models to heavily overfit, optimizing the ratio (even if there are more parameters than there are tokens) is critical. Therefore, I required another corpus of data that I did not have access to.
+
+As a result, I turned to the strategy of using *Off Policy, Sequence Level Knowledge Distillation* from a larger model (Llama 3.2 1B). First, I finetuned the model on the corpus of meditations using [Unsloth AI's notebook](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.2_(1B_and_3B)-Conversational.ipynb).
+Then, I used it to generate approximately 122k tokens of synthetic data over 100 iterations asking it common stoic prompts. I used Google Colab's inbuilt Tesla T4 GPU to run this data generation, which took about 1.5 hours. I do not match logits or trajectories, only generated text; therefore, this approach also pulls elements from instruct-style distillation or offline student-teacher SFT methods. 
+Note: I did not run it from the project directory due to my lack of GPU configuration; however, the adapted notebook has been included.
+
+One critical limitation in this approach is the inefficacy of my prompt library: I did not explicitly instruct the model to focus on _Meditations_ by Marcus Aurelius, enabling the model to hallucinate and pull from various sources of data.
+Future iterations of AureliusGPT-Torch will account for this problem thoroughly to avoid brittle LoRA memory being further defeated, or move to another fine tuning/RL based technique. The core point of this corpus of data was to source philosophical, 
+Stoic or Stoic adjacent data to ensure better generation quality.
+
+My preprocessing logic between AureliusGPT-Torch and AureliusGPT is the same; I rely on Greek transliterations, Unicode normalization, regex patterns, and special "<BEGIN>", "<END>", and "<PAD>" (at training time) tokens.
+I take a similar approach for my preprocessing of _Meditations_ corpus to feed into LoRA; to avoid confusion between Llama 3.2 1B's internal tokens and mine, I avoid adding them in, instead semantically replacing them after the corpus has been generated.
+
+I added the Meditations data twice to the final corpus to weight its significance, especially given its lower token count and the lower quality of synthetic data. My training split was 80% of the pure Meditations data and all the synthetic data; my validation split was 20% of the pure Meditations data.
+
+## Architecture
+
+### Overview
+I use PyTorch's weight initialization (as opposed to a Gaussian process W ~ 𝒩(0, 0.02) for my manual AureliusGPT). I rely on the Transformer architecture from Attention is All You Need (Vaswani et al.) in my implementation.
+Given the small scale of this project, all training (except for the Llama 3.2 1B LoRA for synthetic data) was conducted on a local CPU, and was sequential (not concurrent, hence my lack of ThreadPool or ProcessPoolExecutor).
+I rely on a PostNorm, 6 Transformer block architecture for AureliusGPT-Torch of the format
+
+GPT -> {TransformerBlock -> TransformerBlock -> TransformerBlock...}x6 -> logits
+TransformerBlock -> {AttentionBlock -> Add/Norm -> FFN -> Add/Norm}
+AttentionBlock -> {AttentionHead + Concatenation}
+
+### Embeddings
+I use sinusoidal positional embeddings; given my low parameter budget, I thought it was economical to avoid learning them. My core embedding matrix E is learned.
+
+### Attention
+I use multihead attention in this project, as well as a causal attention mask.
+
+### LayerNorm
+I use PyTorch's inbuilt LayerNorm rather than my own implementation.
+
+
+## Training
+
+### Overview
+
+As mentioned earlier, I incorporate a train/val split of 80:20. I also compute the training and validation loss, as well as the gradient normalization to track overfitting. 
+num_epochs=50 is too high a number; you can using matplotlib graph the relevant losses and identify signs of overfitting during experiments. There is a clear route to On Policy Distillation
+and feedback loops/PPO, RLHF, or RLAIF; there is a method in the original Unsloth AI notebook to save the LoRA weights of the _Meditations_ tuned teacher model to HuggingFace, which can be leveraged as a teacher 
+model in the future. 
+
+### Inference 
+
+While I rely on Off Policy, Sequence Level Knowledge Distillation with Llama 3.2 1B as outlined in my Data section, there is a clear route to implementing Best of N Sampling through concurrent model generations and rankings.
+This can again rely on the finetuned Llama 3.2 1B model or any comparable instruct model on its level.
+
+This model, once fully completed, will be put on HuggingFace and hosted on my website (tarush.ai/aureliusgpt-torch). 
+
+## Future Work on AureliusGPT
+
+### Synthetic Data Generation / Teacher Model
+
+Currently, my LoRA'd Llama 3.2 1B model is run in an ipynb on a Tesla T4 GPU. A future version will integrate the LoRA and synthetic data generation, and relevant GPU plumbing, into the scope of this project. 
+
+Additionally, the universal "teacher/Stoic justifier" model will be the adapted Llama 3.2 1B model, deviating from the OpenAI Chat Completions API GPT-4o approach. 
+
+### Model Visibility on HuggingFace
+
+In a future version, the fitting and accuracy optimized (see Overfitting Tracking / Adaptable Training) files of Llama 3.2 1B's LoRA weights and AureliusGPT-Torch will be loaded onto HuggingFace for fast import and inference.
+
+### Top N Sampling
+
+After the teacher model is converted to Llama 3.2 1B, I will implement config editable concurrent AureliusGPT generation and top N sampling to ensure the highest quality result.
+
+### Overfitting Tracking / Adaptable Training
+
+Currently, training highlights training and validation loss, as well as gradient normalization based on the train/test split to identify overfitting. In a future version, this will be tracked in an easy-to-interpret, modular plot for the user for ease of training.
+
+Beyond weight tuning, config.py will be helpfully automatically adjusted in a future version, changing the learning rate, number of epochs, batch size, and other aspects of training. Therefore, after running n training cycles on its own, the model will iteratively improve its training performance so minimal training optimization is required. 
+
+### Model Upscale
+
+A future project will use the LoRA'd Llama 3.2 1B model to generate signifcantly more parameters of stoic adjacent text, as well as utilizing the works of Zeno, Epictetus, and other famous Stoic thinkers, to build either a Transformer or an MoE model ("Epictetus": _____, "Zeno": _______, "Aurelius": _______) called "StoicDiscourseLM". This will incorporate many elements (including preprocessing functionality) of AureliusGPT but will also be a unique project.
+
+
+## Sample Outputs
+
+### Overview
+
+Some malformed characters and spacing artifacts are due to SentencePiece segmentation and Unicode normalization interactions at small scale, not post-processing errors. With an increased vocab size and optimized Unicode, these errors would not persist.
+
+The config file used to generate these results is on GitHub, but here are the relevant stats:
+```
+vocab_size = 4000
+
+num_blocks = 6
+d_model = 64
+h = 4
+d_head = 16
+d_ff = 64
+
+#training time
+batch_size = 512
+num_epochs = 50
+max_seq_length = 512
+lr = 1e-3
+
+#runtime
+max_tokens = 200
+temperature = 0.8
+justification_model = "gpt-4"
+argmax = False     
+```
+
+
+### Sample Output 1: "The world "
+
+AureliusGPT
+
+A model trained on Meditations by Marcus Aurelius.
+
+User: The world
+
+AURELIUS: for us, when therefore even in all shall never be eth, to have no thought to believe not say unto me. If it will this kind of them to strive and by right, is, what respect and that, or wise as a present station, yea it be with such a man, and opportunity to thyself unto himself from all these ever to be bound to be hindered; and to his own proper natural dreams I cannot be his mind. Thou must needs any man. Be get ative faculty, and see whatsoever it is of a kind of the true and reason, and nature of the general condition of the atre and affections to live according to any man ' s rational substance, a man doth subject to live that seemed nor bad and of comm ent orderly man ' s apparent that is nothing more moderate cond escend ing to the mind and so towards him that was made for a man might comp are leisure to teach with ance of an honest! This ph ed ing to live, and knowledge, only.  
+
+ 
+Stoic Justification Agent: 
+Score: 60/100
+
+Accuracy Report:
+The model's generation has semantic, linguistic and Stoic-based connections with Marcus Aurelius' Meditations, but also presents a number of linguistic inaccuracies and inconsistencies. The text maintains a solemn and introspective tone akin to the Stoic philosophy depicted in Meditations, with phrases such as "rational substance," "natural dreams," and "general condition," all of which could plausibly reflect Aurelius’ contemplations on life, existence, and virtue.
+
+Some sentences or phrases, like "Be get ative faculty, and see whatsoever it is of a kind of the true and reason", seem to bear traces of Stoic themes, but their meaning is encumbered by grammatical errors and awkward wording, lowering the overall score.
+
+The phrase "opportunity to thyself unto himself" seems to echo Stoic teachings about self-reliance and introspection, while "man doth subject to live that seemed nor bad and of comm ent orderly man" and "a man might comp are leisure to teach with ance of an honest!" are hard to parse linguistically and thematically, bringing down the score further.
+
+Overall, while the output bears semblances to the Stoic philosophy and style of Marcus Aurelius, the linguistic inaccuracies and incoherencies limit the model's score. Improvements could be made through refining the model's grasp of grammatical structures and semantic context.
+
+---
+
+### Sample Output 2: "Death "
+
+User: Death
+
+AURELIUS: if one that which ever not to some common good things con ven eth ents unto him, and the whole, but as concer t and the generation of what is inf ition of death dwelt with a blow as it with w ounded which is fitting; but a man would it be said to the vulgar apprehend the several of the feet, and solid th? set to the earth of pleasures? an act ate man ' s upon it, every man as a fool too? when matters, and whatsoever else whereof the whole series of blood of imp ure from all things are very truth, if thou wouldst thou dost think un as to the ext ortion again that perfect simplicity. If thou live with a kos mos or to say, and familiarly ue; and to the common good, and what things in very little while, and as that action? and of time? Those things anim ated into the matter of another, what manner of moment. If thou hast thou art now present are that to nature. What is done. Cont rary: is so must all things in regard the end of the common substance, and who otherwise it, and righteousness, and give thank that can live into perform all agree with it, and according to lose a man, and absolutely and what is a reasonable, and the whole world? and is properly. To a man, that knows not seen, is proper unto thee? towards thy flesh; that is but a rational sociable, and rational, towards them. Why should be a flood and in thy sympathy with it may but draw thy thoughts and credit must be hindered from the common good things that are they doth resist ed ing. For all things. And yet they have done, upon every one point, who is the same kind.  
+
+ 
+Stoic Justification Agent: 
+Score: 45/100
+
+Accuracy Report:
+
+The model's response is a complex body of text with a lot of various elements involved. Parts of it seem to be relating to stoicism's thoughts on death, the interconnectedness of all things, living in the moment and accepting our natural progression towards death. Words that fit with Aurelius's meditations like "common good", "rational", "nature", "simplicity", and questions about life's actions and the nature of living beings are included, showing the model's capability to generate stoic-based content in line with the author.
+
+However, the output does lose points for coherency and linguistic accuracy - it seems to be piecing together phrases and ideas without a clear flow or a strong connection to the initial prompt of "Death". The composition is difficult to comprehend fully, with several instances of phrases being cut and jumbled up, affecting the readability and comprehension. This creates a somewhat disjointed narrative that might not offer the user a satisfactory response to their prompt.
+
+In semantic terms, the toy model presents various stoic elements like maintaining virtues, contemplating nature, and contemplating life and death, which were central themes in the meditations of Marcus Aurelius. 
+
+The model also presents some reflections on human nature, societal matters, and philosophy, showing its ability to generate content from the training material. However, these reflections are scattered throughout and don't weave into a coherent message.
+
+Overall, considering this is a toy model and its limitations, the generated output shows some application of the stoic principles as seen in Meditations by Marcus Aurelius, but lacks the polish and coherence of the original text.
+
+---
+
+### Sample Output 3: "Pain is "
+
+AureliusGPT
+
+A model trained on Meditations by Marcus Aurelius.
+
+User: Pain is 
+
+AURELIUS: a common end. Whatsoever is kindly now can be, when the common proper use of all external accidents, as it doth as it be. They cannot be thy pray, doth his own proper and when are now properly be that it were that which above, even the Pl atonic, and upon this fort ified or if a man whosoever of the fig - father against reason with thee, and that common part of conv ulsion. But in matter of another; whether as that rest contented both resumed and with the things that his own operation.  ⁇ eep thyself, and all doth in others maintain.  ⁇ emember my youth s of the mind as in those that is an ext remity soever is a matter of other peculiar property, or fe ite, but for those things, nor the present themselves, and which is, hast taken out of it be the further ance of a work. That which at all things that which if it were the worse than we see with a perpetual change, and yet even with a w its of all things present ' s commonwealth, that which the satisf action of a man make rec an, it. See this is but as so much as in passion and the city, they should good things that thou such a happy man should ) so that thy refuge, is but best, nor, though but that which is to the Gods, is incidental unto that which is fitting; it be not to pass together, that that which is no more than this that is chiefest, and this, fear that is also is benef man as hitherto thou art find s neighbour should not in this doth himself, or some common good or friend ly, and what there is a man ' s nature that which we may be not in all other course of all this end is, or liberty or ators are the universe did he that both of man is an error, what is mere change? or evil. If therefore is the man likewise. Now all, and that is the series and un sp and the case and that which is either lot and make himself from the mind is now, that which is bound to be no hurt, if my possession of the several roll mutation and that which is not in his kind of all his neighbour is now be a man ' s uch is sociably and necessary? or a man, or a series of those that, that unto man may be a man either a man before after a man whose end; but a man, by man, earth; and the common to whom thy thoughts with this place to live long, and conceit and objects, honour, and re ward on ge song to depart avoid or dispersion, therein any man ste ady, thou didst not lust after one that neither with anything, or st y.  ⁇ es; and that which is proper, is sure that which is allotted unto man whose arm ho pes, I think in regard of all anger: whether all this nature, which is most justly be no man ' s turn. To be a man should rest and separation to be in a common earth, and all things, and duration and all things themselves it were not in some man to embrace whatsoever doth send; and that part and from it were transl ated; therefore, and all things that weight y him, and it, temperate, for their wills that are reasonable faculty, and another: from the examples of such a very little pro cure that which is unto her,) are in thy thoughts that is said, before. By so always be no evil; good and una ffect not grieve more fre quent public like one that the reason, is done, which is proper and what things, or pain ful both be performed to live according to represent pt ations of another is, and the use to do nothing against their pleasure, and opinions, but withdraw. Then long.  ⁇ ep id either the world. It is, either reasonable administration s work of an instant ), is end. What then at all hurtful unto the C ord ant; and not to want of all things: and this and how either form and commonwealth, is, or proper to all substances. But if he is it suffice from all kind and so are of those things in them sw pp ly my particular, and all resumed to man may be a very ridiculous as it is sociably and coherence of a man, why should endeavour anything else, and a certain end.  
+
+ 
+Stoic Justification Agent: 
+This output gets a score of 40/100. 
+
+The model seems to have taken some elements from "Meditations", like the discussions of commonality, the state of being, mortality, and reason, which align fairly well with Stoic philosophies. For instance, Marcus Aurelius often contemplates the nature of humans, life, and existence in the universe, as well as postulates on proper rational behavior. 
+
+However, these themes are drastically overshadowed by the model's incoherency. Semantically, it fails slightly because the output fails to follow a single coherent topic. Instead, there are sudden jumps from one idea to another resulting in lack of contextual associations between sentences. For example, the phrases like "the mind is now", "this end is", or "thy refuge, is but best, nor, though but that which is to the gods, is incidental unto" are hard or impossible to comprehend logically.
+
+From a linguistic perspective, the model fails significantly. There's repetitive usage of certain phrases such as "that which is" and "man", which greatly hampers the text's readability. Moreover, the response is littered with disjointed phrases, narrative inconsistencies, grammatical errors, and meaningless symbols, which makes the text challenging to read and comprehend.
+
+In the context of stoic teachings, while the text vaguely alludes to themes of mortality, human nature, and life's ephemity that Marcus Aurelius often touched upon, it fails to do so in a meaningful or insightful manner. The output hardly embodies stoicism's emphasis on logic, self-discipline, and practical wisdom.
+
+In conclusion, while the model has some alignment with the overall themes of "Meditations", the output lacks coherence, linguistic accuracy, and effective conveyance of Stoic philosophy. Significant work is needed to improve on these areas.

config.py

@@ -0,0 +1,27 @@
+import os
+PROJECT_ROOT = os.path.dirname(os.path.abspath(__file__))
+
+#preprocessing
+greekwords = {'ἔμφρων': 'emphron', 'σύμφρων': 'sumphron', 'ὑπέρφρων': 'huperphron', 'Νόμος': 'nomos', 'νέμων': 'nemon', 'ἀξιοπίστως': 'axiopistos', 'θεοφόρητος': 'theophoretos', 'οἰκονομίαν': 'oikonomian', 'τοῦτο': 'touto', 'ἔφερεν': 'eferen', 'αὐτῷ': 'auto', 'εὔμοιρος': 'eumoiros', 'εὐδαιμονία': 'eudaimonia', 'εὐπατρίδαι': 'eupatridai', 'καθότι': 'kathoti', 'κατορθώσεως': 'katorthoseos', 'κόσμος': 'kosmos', 'μέλος': 'melos', 'μέρος': 'meros', 'παρειλήφαμεν': 'pareilephamen', 'συμβαίνειν': 'symbainein', 'τάσις': 'tasis', 'ἀγαθός': 'agathos', 'ἀκτῖνες': 'aktines', 'ἐκτείνεσθαι': 'ekteinesthai', 'δαίμων': 'daimon', 'κατορθώσεις': 'katorthoseis', 'ἀγαθὸς': 'agathos', 'ἀυτῷ': 'auto'}
+
+#tokenization
+vocab_size = 4000
+
+#model
+num_blocks = 6
+d_model = 64
+h = 4
+d_head = d_model // h
+d_ff = 4 * d_model
+
+#training time
+batch_size = 512
+num_epochs = 50
+max_seq_length = 512
+lr = 1e-3
+
+#runtime
+max_tokens = 200
+temperature = 0.8
+justification_model = "gpt-4"
+argmax = False

epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4a1627a6ae4af12b5cc67d4be0e342c930cb8e06d6bfc05d73dc8e1f237ed94
+size 3312111

handler.py

@@ -0,0 +1,111 @@
+import os
+import sys
+import re
+from typing import Dict, List, Any, Union
+
+import torch
+
+# Ensure your repo root wins import resolution (prevents "model" name collisions)
+REPO_ROOT = os.path.dirname(os.path.abspath(__file__))
+if REPO_ROOT not in sys.path:
+    sys.path.insert(0, REPO_ROOT)
+
+from model.model import Transformer
+from model.vocab.tokenizer import Tokenizer
+import config
+
+
+class EndpointHandler:
+    def __init__(self, path: str = ""):
+        # HF passes `path` pointing to the model snapshot directory.
+        # If it is empty, fall back to repo root.
+        self.base_dir = path or REPO_ROOT
+
+        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+        ckpt_path = os.path.join(self.base_dir, "epoch_10.pt")
+        if not os.path.isfile(ckpt_path):
+            raise FileNotFoundError(f"Missing checkpoint at: {ckpt_path}")
+
+        # NOTE: If your Transformer needs constructor args to match training, pass them here.
+        self.model = Transformer().to(self.device)
+
+        ckpt = torch.load(ckpt_path, map_location=self.device)
+
+        # Support either a raw state_dict OR a checkpoint dict
+        if isinstance(ckpt, dict) and "state_dict" in ckpt:
+            state_dict = ckpt["state_dict"]
+        elif isinstance(ckpt, dict) and "model_state_dict" in ckpt:
+            state_dict = ckpt["model_state_dict"]
+        else:
+            state_dict = ckpt
+
+        self.model.load_state_dict(state_dict, strict=True)
+        self.model.eval()
+
+        # If your Tokenizer expects files relative to CWD, it may break in an endpoint.
+        # Best practice is: make Tokenizer load files relative to its own file location,
+        # OR pass absolute paths derived from `self.base_dir`.
+        self.tokenizer = Tokenizer()
+
+    def _last_token_logits(self, model_out: torch.Tensor) -> torch.Tensor:
+        # Handle common shapes
+        if model_out.dim() == 3:      # [B, T, V]
+            return model_out[0, -1, :]
+        if model_out.dim() == 2:      # [T, V]
+            return model_out[-1, :]
+        raise ValueError(f"Unexpected model output shape: {tuple(model_out.shape)}")
+
+    @torch.inference_mode()
+    def _generate_one(self, prompt: str) -> str:
+        encoded = torch.as_tensor(
+            self.tokenizer.encode(prompt),
+            dtype=torch.long,
+            device=self.device,
+        )
+
+        currtoken = ""
+        outputstring = ""
+        countcheck = 0
+
+        while currtoken != "<END>" and countcheck < config.max_tokens:
+            logits = self._last_token_logits(self.model(encoded))
+
+            if config.argmax:
+                next_id = int(torch.argmax(logits).item())
+            else:
+                probs = torch.softmax(logits / config.temperature, dim=-1)
+                next_id = int(torch.multinomial(probs, num_samples=1).item())
+
+            currtoken = self.tokenizer.decode([next_id]).strip()
+
+            if re.match(r"^[.,!?;:]", currtoken):
+                if outputstring.endswith(" "):
+                    outputstring = outputstring[:-1]
+                outputstring += currtoken + " "
+            else:
+                outputstring += currtoken + " "
+
+            encoded = torch.cat(
+                [encoded, torch.tensor([next_id], dtype=torch.long, device=self.device)],
+                dim=0,
+            )
+            if encoded.numel() > config.max_seq_length:
+                encoded = encoded[-config.max_seq_length :]
+
+            countcheck += 1
+
+        text = re.sub("<BEGIN>", "\n\n", outputstring)
+        text = re.sub("<END>", "\n\n", text)
+        return "AURELIUS: " + text
+
+    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
+        inputs = data.get("inputs", data)
+
+        if isinstance(inputs, dict):
+            inputs = inputs.get("text", "")
+
+        if isinstance(inputs, list):
+            return [{"generated_text": self._generate_one(str(x))} for x in inputs]
+
+        return [{"generated_text": self._generate_one(str(inputs))}]

model/datageneration/llama321b_meditations_lora_unsloth.py

@@ -0,0 +1,230 @@
+# -*- coding: utf-8 -*-
+"""Copy of Llama3.2_(1B_and_3B)-Conversational.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1oXc-BKgPFJJEtekuq-Iqy8WjtwkWVB6I
+
+To run this, press "*Runtime*" and press "*Run all*" on a **free** Tesla T4 Google Colab instance!
+<div class="align-center">
+<a href="https://unsloth.ai/"><img src="https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png" width="115"></a>
+<a href="https://discord.gg/unsloth"><img src="https://github.com/unslothai/unsloth/raw/main/images/Discord button.png" width="145"></a>
+<a href="https://docs.unsloth.ai/"><img src="https://github.com/unslothai/unsloth/blob/main/images/documentation%20green%20button.png?raw=true" width="125"></a></a> Join Discord if you need help + ⭐ <i>Star us on <a href="https://github.com/unslothai/unsloth">Github</a> </i> ⭐
+</div>
+
+To install Unsloth your local device, follow [our guide](https://docs.unsloth.ai/get-started/install-and-update). This notebook is licensed [LGPL-3.0](https://github.com/unslothai/notebooks?tab=LGPL-3.0-1-ov-file#readme).
+
+You will learn how to do [data prep](#Data), how to [train](#Train), how to [run the model](#Inference), & [how to save it](#Save)
+
+### Installation
+"""
+
+# Commented out IPython magic to ensure Python compatibility.
+# %%capture
+# import os, re
+# if "COLAB_" not in "".join(os.environ.keys()):
+#     !pip install unsloth
+# else:
+#     # Do this only in Colab notebooks! Otherwise use pip install unsloth
+#     import torch; v = re.match(r"[0-9]{1,}\.[0-9]{1,}", str(torch.__version__)).group(0)
+#     xformers = "xformers==" + ("0.0.33.post1" if v=="2.9" else "0.0.32.post2" if v=="2.8" else "0.0.29.post3")
+#     !pip install --no-deps bitsandbytes accelerate {xformers} peft trl triton cut_cross_entropy unsloth_zoo
+#     !pip install sentencepiece protobuf "datasets==4.3.0" "huggingface_hub>=0.34.0" hf_transfer
+#     !pip install --no-deps unsloth
+# !pip install transformers==4.56.2
+# !pip install --no-deps trl==0.22.2
+
+"""### Unsloth"""
+
+from unsloth import FastLanguageModel
+import torch
+max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
+dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
+load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.
+
+# 4bit pre quantized models we support for 4x faster downloading + no OOMs.
+fourbit_models = [
+    "unsloth/Meta-Llama-3.1-8B-bnb-4bit",      # Llama-3.1 2x faster
+    "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit",
+    "unsloth/Meta-Llama-3.1-70B-bnb-4bit",
+    "unsloth/Meta-Llama-3.1-405B-bnb-4bit",    # 4bit for 405b!
+    "unsloth/Mistral-Small-Instruct-2409",     # Mistral 22b 2x faster!
+    "unsloth/mistral-7b-instruct-v0.3-bnb-4bit",
+    "unsloth/Phi-3.5-mini-instruct",           # Phi-3.5 2x faster!
+    "unsloth/Phi-3-medium-4k-instruct",
+    "unsloth/gemma-2-9b-bnb-4bit",
+    "unsloth/gemma-2-27b-bnb-4bit",            # Gemma 2x faster!
+
+    "unsloth/Llama-3.2-1B-bnb-4bit",           # NEW! Llama 3.2 models
+    "unsloth/Llama-3.2-1B-Instruct-bnb-4bit",
+    "unsloth/Llama-3.2-3B-bnb-4bit",
+    "unsloth/Llama-3.2-3B-Instruct-bnb-4bit",
+
+    "unsloth/Llama-3.3-70B-Instruct-bnb-4bit" # NEW! Llama 3.3 70B!
+] # More models at https://huggingface.co/unsloth
+
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name = "unsloth/Llama-3.2-1B-Instruct",
+    max_seq_length = max_seq_length,
+    dtype = dtype,
+    load_in_4bit = load_in_4bit,
+    # token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
+)
+
+"""We now add LoRA adapters so we only need to update 1 to 10% of all parameters!"""
+
+model = FastLanguageModel.get_peft_model(
+    model,
+    r = 16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
+    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
+                      "gate_proj", "up_proj", "down_proj",],
+    lora_alpha = 16,
+    lora_dropout = 0, # Supports any, but = 0 is optimized
+    bias = "none",    # Supports any, but = "none" is optimized
+    # [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes!
+    use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context
+    random_state = 3407,
+    use_rslora = False,  # We support rank stabilized LoRA
+    loftq_config = None, # And LoftQ
+)
+
+"""<a name="Data"></a>
+### Data Prep
+We now use the `Llama-3.1` format for conversation style finetunes. NOTE: THIS HAS BEEN CHANGED TO THE CORPUS OF MEDITATIONS BY MARCUS AURELIUS (89k TOKENS).
+"""
+
+from datasets import Dataset
+
+with open("input.txt", "r", encoding="utf-8") as f:
+    text = f.read()
+
+# Split into chunks of ~500 chars for training samples
+chunks = [text[i:i+500] for i in range(0, len(text), 500)]
+
+# Format as "Chat" data
+# We trick the model: User asks for wisdom -> Assistant gives a quote.
+data = []
+for chunk in chunks:
+    data.append([
+        {"role": "user", "content": "Write a stoic meditation."},
+        {"role": "assistant", "content": chunk}
+    ])
+
+# Convert to HF Dataset
+dataset = Dataset.from_dict({"conversations": data})
+
+# Apply Unsloth's formatting
+from unsloth.chat_templates import get_chat_template
+tokenizer = get_chat_template(tokenizer, chat_template = "llama-3.1")
+
+def formatting_prompts_func(examples):
+    convos = examples["conversations"]
+    texts = [tokenizer.apply_chat_template(convo, tokenize = False, add_generation_prompt = False) for convo in convos]
+    return { "text" : texts }
+
+dataset = dataset.map(formatting_prompts_func, batched = True)
+
+"""<a name="Train"></a>
+### Train the model
+Now let's train our model. We do 60 steps to speed things up, but you can set `num_train_epochs=1` for a full run, and turn off `max_steps=None`. We also support TRL's `DPOTrainer`!
+"""
+
+from trl import SFTConfig, SFTTrainer
+from transformers import DataCollatorForSeq2Seq
+trainer = SFTTrainer(
+    model = model,
+    tokenizer = tokenizer,
+    train_dataset = dataset,
+    dataset_text_field = "text",
+    max_seq_length = max_seq_length,
+    data_collator = DataCollatorForSeq2Seq(tokenizer = tokenizer),
+    packing = False, # Can make training 5x faster for short sequences.
+    args = SFTConfig(
+        per_device_train_batch_size = 2,
+        gradient_accumulation_steps = 4,
+        warmup_steps = 5,
+        # num_train_epochs = 1, # Set this for 1 full training run.
+        max_steps = 60,
+        learning_rate = 2e-4,
+        logging_steps = 1,
+        optim = "adamw_8bit",
+        weight_decay = 0.001,
+        lr_scheduler_type = "linear",
+        seed = 3407,
+        output_dir = "outputs",
+        report_to = "none", # Use TrackIO/WandB etc
+    ),
+)
+
+# @title Show current memory stats
+gpu_stats = torch.cuda.get_device_properties(0)
+start_gpu_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3)
+max_memory = round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3)
+print(f"GPU = {gpu_stats.name}. Max memory = {max_memory} GB.")
+print(f"{start_gpu_memory} GB of memory reserved.")
+
+trainer_stats = trainer.train()
+
+# @title Show final memory and time stats
+used_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3)
+used_memory_for_lora = round(used_memory - start_gpu_memory, 3)
+used_percentage = round(used_memory / max_memory * 100, 3)
+lora_percentage = round(used_memory_for_lora / max_memory * 100, 3)
+print(f"{trainer_stats.metrics['train_runtime']} seconds used for training.")
+print(
+    f"{round(trainer_stats.metrics['train_runtime']/60, 2)} minutes used for training."
+)
+print(f"Peak reserved memory = {used_memory} GB.")
+print(f"Peak reserved memory for training = {used_memory_for_lora} GB.")
+print(f"Peak reserved memory % of max memory = {used_percentage} %.")
+print(f"Peak reserved memory for training % of max memory = {lora_percentage} %.")
+
+"""Now, we generate 100k parameters of Meditations adjacent synthetic data on the LoRA Llama 3.2 1B instruct."""
+
+
+
+FastLanguageModel.for_inference(model)
+import torch
+
+#initial prompts
+prompts = [
+    "On the shortness of life", "On dealing with anger", "On the nature of the universe",
+    "How to face death", "The duty of a leader", "On logic and reason",
+    "On the behavior of others", "Why fame is meaningless", "On finding peace",
+    "On the passage of time", "The virtue of silence", "Endurance of pain"
+]
+
+generated_data = ""
+
+for i in range(100):
+    inputs = tokenizer(prompts, return_tensors = "pt", padding=True).to("cuda") #parallelism for GPU
+
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens = 256,
+        temperature = 1.2, #higher temperature is more creative; better for training data
+        min_p = 0.1
+    )
+
+    decoded_texts = tokenizer.batch_decode(outputs, skip_special_tokens=True)
+
+    for text in decoded_texts:
+        # Llama has a tendency to repeat the prompt, so we remove that; strip whitespace
+        clean_content = text.replace("Write a stoic meditation.", "").strip()
+
+        if len(clean_content) > 20:
+            tagged_entry = f"<BEGIN> {clean_content} <END>\n\n" #introduces original tags back into the synthetic data
+            generated_data += tagged_entry
+
+    if i % 10 == 0:
+        print(f"Batch {i}/100 complete; total_chars = {len(generated_data)}")
+
+output_filename = "synthetic_meditations.txt"
+
+#save it
+with open(output_filename, "w") as f:
+    f.write(generated_data)
+
+#if desired, you can view the original notebook for instructions on saving the LoRA adapters or full 16bit GGUF to HF.
+#You could then use this as the justification functionality or for On Policy Distillation (not Off Policy/Sequence Level Knowledge Distillation as is implemented here).

model/datageneration/syntheticformatting.py

@@ -0,0 +1,42 @@
+import re, os, unicodedata
+from config import PROJECT_ROOT, greekwords
+
+class SyntheticFormatter:
+    def __init__(self):
+        pass
+
+    def process(self, text):
+        meditations = unicodedata.normalize("NFC", text)
+
+        try:
+            startindex = meditations.index("THE FIRST BOOK")
+            meditations = meditations[startindex:]
+            endindex = meditations.index("APPENDIX")
+            meditations = meditations[:endindex] 
+        except ValueError:
+            pass
+
+        book_name = r"THE\s+[A-Z]+\s+BOOK\s+[IVXLCDM]+\."
+        section_name = r"\n\n[IVXLCDM]+\. "
+        underline = r"[_]+"
+        newline_in_sentence = r"(?<!\n)\n(?!\n)"
+
+        for key, value in greekwords.items():
+            meditations = meditations.replace(key, value)
+
+        meditations = re.sub(newline_in_sentence, " ", meditations)
+        meditations = re.sub(book_name, "\n\n", meditations)
+        meditations = re.sub(section_name, "\n\n", meditations)
+        meditations = re.sub(underline, "", meditations)
+        meditations = re.sub(r'\n{3,}', '\n\n', meditations)
+        
+        meditations = meditations.strip()
+
+        final_sections = [s.strip() for s in meditations.split('\n\n') if s.strip()]
+        
+        return meditations, final_sections
+
+    def split_into_sentences(self, text: str) -> list[str]:
+        sentences = re.split(r'(?<=[.!?])\s+', text)
+        return [s.strip() for s in sentences if s.strip()]
+

model/model.py

@@ -0,0 +1,104 @@
+import numpy as np
+import torch
+import torch.nn as nn
+from config import num_blocks, vocab_size, d_model, h, d_head, d_ff, max_seq_length
+
+#main transformer
+
+class Util:
+    def sinusoidal(self):
+        PE = np.zeros((max_seq_length, d_model))
+        
+        for pos in range(max_seq_length):
+            for i in range(0, d_model, 2):
+                div_term = 10000 ** (i / d_model)
+                PE[pos, i] = np.sin(pos / div_term)
+                if i + 1 < d_model:
+                    PE[pos, i + 1] = np.cos(pos / div_term)
+                    
+        return PE
+
+class Transformer(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.blocks = nn.ModuleList([TransformerBlock() for i in range(num_blocks)])
+        self.embeddings = nn.Embedding(num_embeddings=vocab_size, embedding_dim=d_model)
+        util = Util()
+        self.positionals = util.sinusoidal()
+        self.linear = nn.Linear(d_model, vocab_size)
+
+    def forward(self, X):
+        embeddings = self.embeddings(X)
+        positionals = torch.tensor(self.positionals[:X.shape[0]]).float() 
+        embeddings = embeddings + positionals
+
+        for block in self.blocks:
+            embeddings = block(embeddings)
+
+        return self.linear(embeddings)
+
+class TransformerBlock(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.attentionblock = AttentionBlock()
+        self.layernorm = LayerNorm()
+        self.ffn = FFN()
+        self.layernorm2 = LayerNorm()
+    
+    def forward(self, X):
+        X = self.layernorm(X + self.attentionblock(X))
+        X = self.layernorm2(X + self.ffn(X))
+        return X
+
+#attention
+class AttentionBlock(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.attentionheads = nn.ModuleList([AttentionHead() for i in range(h)])
+        self.Wo = nn.Linear(d_model, d_model)
+
+    def forward(self, X):
+        headoutputs = [head(X) for head in self.attentionheads]
+        MHA = torch.cat(headoutputs, dim=-1)
+        return self.Wo(MHA)
+    
+
+class AttentionHead(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.queries = nn.Linear(d_model, d_head, bias=False)
+        self.keys = nn.Linear(d_model, d_head, bias=False)
+        self.values = nn.Linear(d_model, d_head, bias=False)
+        
+    def forward(self, X):
+        Q = self.queries(X)
+        K = self.keys(X)
+        V = self.values(X)
+
+        scores = Q @ K.T
+        scores /= (d_head ** 0.5)
+        mask = torch.tril(torch.ones(X.shape[0], X.shape[0]))
+        scores = scores.masked_fill(mask == 0, float('-inf'))
+        attention = torch.softmax(scores, dim=-1)  
+        return attention @ V      
+
+#adding / prenorm
+class LayerNorm(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.norm = nn.LayerNorm(d_model)
+    def forward(self, X):
+        return self.norm(X)
+
+#ffn
+class FFN(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.net = nn.Sequential(
+            nn.Linear(d_model, d_ff),
+            nn.ReLU(), 
+            nn.Linear(d_ff, d_model)
+        )
+
+    def forward(self, X):
+        return self.net(X)

model/run.py

@@ -0,0 +1,105 @@
+import os, torch, re
+import torch.nn as nn
+from config import PROJECT_ROOT, num_epochs, max_tokens, temperature, max_seq_length, justification_model, argmax
+from model.model import Transformer
+from model.vocab.tokenizer import Tokenizer
+from model.vocab.preprocess import Preprocessor
+from openai import OpenAI
+from dotenv import load_dotenv
+
+load_dotenv()
+
+class Run:
+    def __init__(self):
+        self.tokenizer = Tokenizer()
+        self.model = Transformer()
+        self.path = os.path.join(PROJECT_ROOT, "data", f"epoch_10.pt")
+        self.model.load_state_dict(torch.load(self.path)) 
+        self.model.eval()
+
+    def input(self):
+        input_query = input("User: ")
+        return input_query
+
+    def run(self, input_query):
+        
+        if input_query == "":
+            return "INTERACTION_COMPLETE"
+
+        print("\n")
+        encoded = self.tokenizer.encode(input_query)
+        encoded = torch.LongTensor(encoded)
+        currtoken = ""
+        outputstring = ""
+        countcheck = 0
+        while currtoken != "<END>" and countcheck < max_tokens:
+            with torch.no_grad(): 
+                currtoken = self.model(encoded)
+                last_token_logits = currtoken[-1, :] 
+
+
+                if argmax:
+                    predictions = torch.argmax(last_token_logits).item()
+                else:
+                    logits = last_token_logits / temperature
+                    probs = torch.softmax(logits, dim=-1)
+                    predictions = torch.multinomial(probs, num_samples=1).item()
+
+                currtoken = self.tokenizer.decode([predictions]).strip()
+                if re.match(r'^[.,!?;:]', currtoken):
+                    if outputstring.endswith(" "):
+                        outputstring = outputstring[:-1]
+                    outputstring += currtoken + " "
+                else:
+                    outputstring += currtoken + " "
+                next_token = torch.tensor([predictions], dtype=torch.long)
+                encoded = torch.cat((encoded, next_token), dim=0)
+                if encoded.shape[0] > max_seq_length:
+                    encoded = encoded[-max_seq_length:]
+        currtoken = predictions
+
+        return input_query, outputstring
+    
+    def postprocess(self, text):
+        text = re.sub("<BEGIN>", "\n\n", text)
+        text = re.sub("<END>", "\n\n", text)
+        return "AURELIUS: " + text
+
+    def justify(self, text):
+        client = OpenAI()
+
+        prompt = "Your responsibility is to justify the output of a toy model fitted on Meditations by Marcus Aurelius. Please read the user's prompt, the model's response, and give the model a score out of 100 of prediction given it's status as a toy model. Generate a short report of its accuracy, identifying potential semantic, linguistic, and Stoic-meaning based connections in the generation."
+
+        response = client.chat.completions.create(
+            model=justification_model,
+            messages=[
+                {"role": "system", "content": prompt},
+                {"role": "user", "content": text}
+            ]
+        )
+
+        print("Stoic Justification Agent: \n" + response.choices[0].message.content)
+
+    def main(self):
+        print("\nAureliusGPT\n")
+        print("A model trained on Meditations by Marcus Aurelius.\n\n")
+
+        while True:
+            print("\nPress 'Enter' to stop the conversation.")
+            user_input = self.input()
+
+            if user_input == "":
+                break
+
+            ran = self.run(user_input)
+            outputstring = ran[1]
+            postprocessed = self.postprocess(outputstring)
+            print(postprocessed)
+            print(self.justify("Prompt: " + ran[0] + "\n\n" + postprocessed))
+
+
+
+if __name__ == "__main__":  
+    run = Run()
+    run.main()
+    

model/train.py

@@ -0,0 +1,145 @@
+import os, torch, torch.nn as nn
+import random
+from model.datageneration.syntheticformatting import SyntheticFormatter
+from model.vocab.preprocess import Preprocessor
+from model.vocab.tokenizer import Tokenizer
+from model.model import Transformer
+from config import PROJECT_ROOT, batch_size, num_epochs, lr
+
+class Train:
+    def __init__(self):
+        self.formatter = SyntheticFormatter()
+        self.pre = Preprocessor()
+        self.tokenizer = Tokenizer()
+        self.model = Transformer()
+        self.loss = nn.CrossEntropyLoss()
+        self.optimizer = torch.optim.Adam(self.model.parameters(), lr)
+        self.weight_path = os.path.join(PROJECT_ROOT, "data", "weights.pt")
+        
+        # Placeholders for data
+        self.train_tokenized = []
+        self.val_tokenized = []
+
+    def generate_synthetic_meditations(self):
+        with open(os.path.join(PROJECT_ROOT, "meditations.txt"), "r") as f:
+            meditations = f.read()
+        processed = self.formatter.process(meditations)
+
+        with open(os.path.join(PROJECT_ROOT, "meditationssynthetic.txt"), "w") as f:
+            f.write(processed[0])
+
+        #Here is where we would use processed meditations (no <BEGIN> or <END> tags)
+        #to run Unsloth LoRA on LLAMA 3.2 1B to generate 100k parameters of synthetic Meditations
+        #adjacent stoic text. The adapted Colab Notebook is at model/llama321b_meditations_lora_unsloth.py
+
+    def process_and_train_tokenizer(self):
+        with open(os.path.join(PROJECT_ROOT, "meditations.txt"), "r") as f:
+            meditations = f.read()
+        
+        meditations_synthetic = ""
+        if os.path.exists(os.path.join(PROJECT_ROOT, "meditationssynthetic.txt")):
+            with open(os.path.join(PROJECT_ROOT, "meditationssynthetic.txt"), "r") as f:
+                meditations_synthetic = f.read()
+
+        processed_pure = self.pre.process(meditations)
+        processed_synthetic = self.pre.process(meditations_synthetic)
+
+        processed_overall_sentences = processed_pure[1] + processed_synthetic[1]
+        self.tokenizer.train(processed_overall_sentences)
+        
+        self.trained, self.val = self.train_val(processed_pure[0], processed_synthetic[0])
+        self.train_tokenized = self.tokenizer.encode(self.trained)
+        self.val_tokenized = self.tokenizer.encode(self.val)
+
+    def train_val(self, pure, synthetic):
+        eighty = 4 * len(pure) // 5
+        val = pure[eighty:]
+        train_pure = pure[:eighty]
+        trained = train_pure + train_pure + synthetic
+        return trained, val
+
+    def seperate_into_batches(self, token_list):
+        seperated = []
+        for i in range(0, len(token_list), batch_size):
+            chunked = token_list[i:i+batch_size]
+            padding_number = batch_size - len(chunked)
+            if padding_number > 0:
+                chunked.extend([self.tokenizer.encode("<PAD>")[0]] * padding_number)
+            seperated.append(chunked)
+        return seperated
+
+    def train_batch(self, batch, print_loss=False):
+        batch = torch.LongTensor(batch)
+        X = batch[0:-1]
+        Y = batch[1:]
+        
+        logits = self.model.forward(X)
+        ce_loss = self.loss(logits.view(-1, logits.size(-1)), Y.view(-1))
+        self.optimizer.zero_grad()
+        ce_loss.backward()
+
+        if print_loss:
+            print("CE loss is: " + str(ce_loss))
+
+        #calculating gradient norm for train/val overfitting check
+        total_norm = 0.0
+        for p in self.model.parameters():
+            if p.grad is not None:
+                param_norm = p.grad.data.norm(2)
+                total_norm += param_norm.item() ** 2
+        total_norm = total_norm ** 0.5
+
+        self.optimizer.step()
+        return ce_loss.item(), total_norm
+
+    def evaluate(self, val_batches):
+        self.model.eval()
+        total_loss = 0
+        with torch.no_grad():
+            for batch in val_batches:
+                batch_tensor = torch.LongTensor(batch)
+                X = batch_tensor[0:-1]
+                Y = batch_tensor[1:]
+                logits = self.model.forward(X)
+                loss = self.loss(logits, Y)
+                total_loss += loss.item()
+        
+        self.model.train()
+        return total_loss / len(val_batches) if val_batches else 0
+
+    def train(self):
+        self.process_and_train_tokenizer()
+
+        train_batches = self.seperate_into_batches(self.train_tokenized)
+        val_batches = self.seperate_into_batches(self.val_tokenized)    
+        
+        for i in range(num_epochs):
+            random.shuffle(train_batches)
+
+            epoch_loss = 0
+            epoch_grad_norm = 0
+
+            for j, batch in enumerate(train_batches):
+                loss, grad_norm = self.train_batch(batch)
+                epoch_loss += loss
+                epoch_grad_norm += grad_norm
+                
+                if j % 500 == 0 and j > 0:
+                    print(f"  Batch {j}; Loss: {loss:.4f}; Grad Norm: {grad_norm:.4f}")
+
+
+            avg_train_loss = epoch_loss / len(train_batches)
+            avg_grad_norm = epoch_grad_norm / len(train_batches)
+            val_loss = self.evaluate(val_batches)
+
+            print(f"Epoch {i+1}; Train Loss: {avg_train_loss:.4f}; Val Loss: {val_loss:.4f}; Train Grad Norm: {avg_grad_norm:.4f}")
+        
+            if (i+1) % 10 == 0:
+                save_path = os.path.join(PROJECT_ROOT, "data", f"epoch_{i+1}.pt")
+                torch.save(self.model.state_dict(), save_path)
+
+
+if __name__ == "__main__":
+    train = Train()
+    #train.generate_synthetic_meditations() #uncomment to generate synthetic meditations, then comment out the next line
+    train.train()

model/vocab/preprocess.py

@@ -0,0 +1,103 @@
+import re, os, unicodedata
+import sys
+from config import PROJECT_ROOT, greekwords
+
+class Preprocessor:
+    def __init__(self):
+        pass
+
+    def process(self, text):
+        meditations = unicodedata.normalize("NFC", text)
+
+        # Check if the text has the expected structure of the original Meditations
+        is_original = "THE FIRST BOOK" in meditations and "APPENDIX" in meditations
+
+        if is_original:
+            startindex = meditations.index("THE FIRST BOOK")
+            meditations = meditations[startindex:]
+            endindex = meditations.index("APPENDIX")
+            meditations = meditations[:endindex] 
+
+            book_name = r"THE\s+[A-Z]+\s+BOOK\s+[IVXLCDM]+\.\s"
+            section_name = r"\n\n[IVXLCDM]+\. "
+            book_end = r"\n\n\n\n"
+            
+            meditations = re.sub(book_name, "<BEGIN> \n", meditations)
+            meditations = re.sub(book_end, "<END> ", meditations)
+            meditations = re.sub(section_name, "\n <END> \n <BEGIN> \n", meditations)
+            
+            split_pattern = f"{book_name}|{section_name}"
+        else:
+            # Minimal processing for synthetic data
+            split_pattern = r"\n\n"  # Split by paragraphs or double newlines
+
+        underline = r"[_]+"
+        newline_in_sentence = r"(?<!\n)\n(?!\n)"
+
+        meditations = re.sub(underline, "", meditations)
+        meditations = re.sub(newline_in_sentence, " ", meditations)
+
+        for key, value in greekwords.items():
+            meditations = meditations.replace(key, value)
+        
+        raw_sections = re.split(split_pattern, meditations)
+
+        final_sections = []
+        for section in raw_sections:
+            if not section.strip():
+                continue
+            sentences = self.split_into_sentences(section)
+            if sentences:
+                final_sections.extend(sentences)
+        
+        return meditations, final_sections
+
+    def split_into_sentences(self, text: str) -> list[str]:
+        sentences = re.split(r'(?<=[.!?])\s+', text)
+        return [s.strip() for s in sentences if s.strip()]
+
+    def test(self, file):
+        processed = None
+        if file:
+            try:
+                processed = self.process(file)
+            except Exception:
+                print("The processed file is not compliant with preprocess' requirements. Falling back to default file.\n")
+                processed = None
+        if not processed:
+            test_file_path = os.path.join(os.path.dirname(__file__), "preprocess_test.txt")
+            with open(test_file_path, "r") as f:
+                processed = self.process(f.read())
+
+        output_file_path = os.path.join(os.path.dirname(__file__), "preprocess_test_output.txt")
+        with open(output_file_path, "w") as f:
+            f.write(processed[0])
+            
+        print(f"Saved to {output_file_path}.")
+
+if __name__ == "__main__":
+    file = None
+    if len(sys.argv) > 1:
+        test = sys.argv[1]
+        if test != "test":
+            print("Only permitted argument is 'test'; Please try again.")
+            pass
+
+    else:
+        print("Preprocessing logic is wrapped into overall training functionality.")
+        pass
+    
+    if len(sys.argv) > 2:
+        filepath = sys.argv[2]
+        try:
+            with open(filepath, "r") as f:
+                file = f.read()
+        except Exception as e:
+            print("Invalid filepath, falling back to original test.")
+            file = None
+
+    Preprocessor().test(file)
+
+
+
+

model/vocab/preprocess_test.txt

@@ -0,0 +1,21 @@
+
+Introduction text that should be skipped.
+
+THE FIRST BOOK
+
+I. This is the first section of the first book. It has some text.
+
+II. This is the second section. It has more text and a greek word: κόσμος.
+And a newline within the sentence.
+
+THE SECOND BOOK
+
+I. First section of the second book.
+
+II. Second section of the second book.
+
+
+
+APPENDIX
+
+Index and other stuff that should be skipped.

model/vocab/tokenize_test.txt

@@ -0,0 +1,5 @@
+This is a test sentence for the tokenizer.
+It includes some punctuation: Hello, world!
+And maybe a Greek word if the model supports it: κόσμος.
+<BEGIN> Special tokens test <END>
+

model/vocab/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d6e162b2739a133a1adfbe22acde59bca5df13e3bccea8e6e226de81d512eee
+size 303849

model/vocab/tokenizer.py

@@ -0,0 +1,81 @@
+import os, io, sys
+import sentencepiece as spm
+from config import PROJECT_ROOT, vocab_size
+from model.vocab.preprocess import Preprocessor
+
+class Tokenizer:
+    def __init__(self):
+        self.path = os.path.join(PROJECT_ROOT, "tokenizer.model")
+        self.sp = spm.SentencePieceProcessor()
+        if os.path.exists(self.path):
+            self.sp.Load(self.path)
+    
+    def train(self, all_sentences):
+        spm.SentencePieceTrainer.train(
+            sentence_iterator=iter(all_sentences), 
+            model_prefix="data/tokenizer", 
+            model_type="bpe",
+            vocab_size=vocab_size,
+            user_defined_symbols=["<BEGIN>", "<END>", "<PAD>"]
+        )
+        self.sp.load(self.path)
+    
+    def encode(self, text):
+        return self.sp.encode(text)
+
+    def decode(self, ids):
+        return self.sp.decode(ids)
+
+
+    def test(self, file):
+        text_content = None
+        if file:
+            text_content = file
+        
+        if not text_content:
+            test_file_path = os.path.join(os.path.dirname(__file__), "tokenize_test.txt")
+            if os.path.exists(test_file_path):
+                with open(test_file_path, "r") as f:
+                    text_content = f.read()
+            else:
+                print(f"Default test file not found at {test_file_path}")
+                return
+
+        if text_content:
+            try:
+                encoded_ids = self.encode(text_content)
+                decoded_text = self.decode(encoded_ids)
+                
+                output_content = f"Original Text:\n{text_content}\n\nToken IDs:\n{encoded_ids}\n\nDecoded Text:\n{decoded_text}\n"
+                
+                output_file_path = os.path.join(os.path.dirname(__file__), "tokenize_test_output.txt")
+                with open(output_file_path, "w") as f:
+                    f.write(output_content)
+                    
+                print(f"Saved to {output_file_path}.")
+            except Exception as e:
+                print(f"Error during tokenization test: {e}")
+
+if __name__ == "__main__":
+    file = None
+    if len(sys.argv) > 1:
+        test = sys.argv[1]
+        if test != "test":
+            print("Only permitted argument is 'test'; Please try again.")
+            pass
+
+    else:
+        print("Tokenization logic is wrapped into overall training functionality.")
+        pass
+
+    if len(sys.argv) > 2:
+        filepath = sys.argv[2]
+        try:
+            with open(filepath, "r") as f:
+                file = f.read()
+        except Exception as e:
+            print("Invalid filepath, falling back to original test.")
+            file = None
+
+    if len(sys.argv) > 1 and sys.argv[1] == "test":
+        Tokenizer().test(file)

model/vocab/tokenizer.vocab

@@ -0,0 +1,4000 @@
+<unk>	0
+<s>	0
+</s>	0
+<BEGIN>	0
+<END>	0
+<PAD>	0
+th	-0
+▁th	-1
+▁a	-2
+er	-3
+in	-4
+▁o	-5
+at	-6
+nd	-7
+▁w	-8
+▁the	-9
+▁s	-10
+▁i	-11
+re	-12
+▁and	-13
+▁b	-14
+on	-15
+▁m	-16
+▁t	-17
+▁h	-18
+ou	-19
+▁c	-20
+se	-21
+▁of	-22
+▁n	-23
+▁that	-24
+or	-25
+st	-26
+▁to	-27
+en	-28
+▁d	-29
+▁f	-30
+▁p	-31
+▁wh	-32
+ing	-33
+ll	-34
+▁be	-35
+an	-36
+▁is	-37
+▁in	-38
+ed	-39
+▁it	-40
+▁e	-41
+▁u	-42
+le	-43
+ar	-44
+ion	-45
+om	-46
+▁no	-47
+ch	-48
+es	-49
+ith	-50
+is	-51
+▁thou	-52
+ver	-53
+ut	-54
+▁or	-55
+▁as	-56
+ly	-57
+it	-58
+▁l	-59
+▁an	-60
+▁ha	-61
+▁g	-62
+▁re	-63
+ce	-64
+nt	-65
+▁thing	-66
+sel	-67
+▁do	-68
+▁all	-69
+▁con	-70
+as	-71
+▁not	-72
+ve	-73
+ld	-74
+▁sh	-75
+al	-76
+ere	-77
+ct	-78
+▁for	-79
+▁whi	-80
+▁thy	-81
+ther	-82
+▁things	-83
+▁so	-84
+▁with	-85
+▁man	-86
+im	-87
+self	-88
+od	-89
+ow	-90
+▁which	-91
+ay	-92
+so	-93
+▁su	-94
+▁on	-95
+ure	-96
+▁unt	-97
+▁he	-98
+▁unto	-99
+id	-100
+ent	-101
+gh	-102
+▁any	-103
+▁I	-104
+ur	-105
+▁what	-106
+ir	-107
+▁pr	-108
+▁them	-109
+ble	-110
+▁li	-111
+▁they	-112
+▁are	-113
+ind	-114
+pp	-115
+▁nat	-116
+ither	-117
+▁A	-118
+ain	-119
+▁by	-120
+▁but	-121
+et	-122
+ever	-123
+all	-124
+ose	-125
+ation	-126
+▁com	-127
+ter	-128
+▁T	-129
+fe	-130
+ore	-131
+▁al	-132
+▁thee	-133
+▁this	-134
+▁fr	-135
+ust	-136
+▁wor	-137
+ic	-138
+▁tr	-139
+▁Th	-140
+▁one	-141
+ill	-142
+art	-143
+ould	-144
+▁nature	-145
+ess	-146
+am	-147
+▁de	-148
+ul	-149
+▁r	-150
+ght	-151
+▁him	-152
+▁F	-153
+▁go	-154
+▁his	-155
+▁pro	-156
+▁from	-157
+ke	-158
+▁ne	-159
+▁doth	-160
+est	-161
+thing	-162
+em	-163
+▁either	-164
+▁if	-165
+▁other	-166
+▁we	-167
+very	-168
+▁part	-169
+▁For	-170
+ome	-171
+per	-172
+il	-173
+▁who	-174
+ad	-175
+▁there	-176
+soever	-177
+▁se	-178
+▁can	-179
+▁k	-180
+▁af	-181
+pe	-182
+▁st	-183
+ity	-184
+▁good	-185
+able	-186
+▁thyself	-187
+▁have	-188
+▁such	-189
+▁those	-190
+▁W	-191
+us	-192
+ge	-193
+ions	-194
+▁their	-195
+cc	-196
+ous	-197
+▁comm	-198
+▁up	-199
+▁then	-200
+▁cont	-201
+▁will	-202
+▁ex	-203
+▁may	-204
+▁tru	-205
+▁world	-206
+ason	-207
+▁at	-208
+▁un	-209
+qu	-210
+▁di	-211
+wn	-212
+our	-213
+▁pre	-214
+▁some	-215
+▁must	-216
+▁y	-217
+▁dis	-218
+ri	-219
+▁should	-220
+▁me	-221
+▁And	-222
+▁happ	-223
+ame	-224
+▁also	-225
+▁Wh	-226
+▁ch	-227
+to	-228
+ong	-229
+▁my	-230
+▁B	-231
+and	-232
+▁mind	-233
+▁when	-234
+▁v	-235
+ard	-236
+ight	-237
+▁how	-238
+▁reason	-239
+ot	-240
+ine	-241
+▁app	-242
+▁shall	-243
+are	-244
+ves	-245
+fore	-246
+out	-247
+▁whatsoever	-248
+ake	-249
+▁le	-250
+▁more	-251
+os	-252
+fect	-253
+▁upon	-254
+▁ag	-255
+▁(	-256
+ime	-257
+▁life	-258
+▁were	-259
+▁these	-260
+▁cons	-261
+▁hath	-262
+oth	-263
+▁acc	-264
+▁ar	-265
+▁own	-266
+ci	-267
+ate	-268
+▁her	-269
+nder	-270
+▁common	-271
+▁ev	-272
+▁H	-273
+ord	-274
+▁men	-275
+▁proper	-276
+ra	-277
+▁many	-278
+▁time	-279
+ire	-280
+sent	-281
+ff	-282
+ber	-283
+▁But	-284
+▁every	-285
+▁art	-286
+▁ob	-287
+▁end	-288
+▁j	-289
+▁act	-290
+alt	-291
+▁was	-292
+other	-293
+▁itself	-294
+▁both	-295
+▁than	-296
+end	-297
+ive	-298
+ance	-299
+▁now	-300
+▁himself	-301
+▁per	-302
+▁anything	-303
+ect	-304
+▁shalt	-305
+▁ab	-306
+▁S	-307
+▁sub	-308
+▁happen	-309
+ath	-310
+ant	-311
+ies	-312
+▁nor	-313
+ak	-314
+ary	-315
+selves	-316
+ally	-317
+reat	-318
+▁yet	-319
+▁into	-320
+▁kind	-321
+▁present	-322
+au	-323
+▁G	-324
+ast	-325
+uch	-326
+▁same	-327
+ence	-328
+▁What	-329
+▁themselves	-330
+one	-331
+▁sou	-332
+atter	-333
+▁again	-334
+▁another	-335
+ist	-336
+▁lo	-337
+▁only	-338
+▁true	-339
+ort	-340
+ness	-341
+▁after	-342
+ward	-343
+ise	-344
+▁im	-345
+ject	-346
+▁very	-347
+ider	-348
+▁did	-349
+▁neither	-350
+▁O	-351
+▁out	-352
+▁much	-353
+▁natur	-354
+▁ple	-355
+ording	-356
+▁under	-357
+pt	-358
+ost	-359
+ree	-360
+ved	-361
+▁hast	-362
+▁therefore	-363
+ice	-364
+rou	-365
+▁ind	-366
+▁according	-367
+ious	-368
+▁ord	-369
+▁N	-370
+ass	-371
+▁God	-372
+um	-373
+tain	-374
+▁bod	-375
+▁use	-376
+eth	-377
+itt	-378
+way	-379
+▁where	-380
+ful	-381
+▁without	-382
+▁consider	-383
+▁soul	-384
+▁matter	-385
+ower	-386
+▁The	-387
+verse	-388
+▁well	-389
+ac	-390
+ener	-391
+▁whole	-392
+▁nothing	-393
+ase	-394
+ause	-395
+▁op	-396
+icul	-397
+▁power	-398
+▁would	-399
+▁affect	-400
+▁pl	-401
+ear	-402
+▁kn	-403
+▁live	-404
+▁gener	-405
+▁am	-406
+▁even	-407
+ud	-408
+▁she	-409
+ss	-410
+ffer	-411
+stand	-412
+ok	-413
+▁ever	-414
+ible	-415
+▁conc	-416
+▁disp	-417
+▁long	-418
+▁make	-419
+▁opin	-420
+icular	-421
+▁never	-422
+▁natural	-423
+▁death	-424
+▁particular	-425
+ens	-426
+wards	-427
+▁know	-428
+▁our	-429
+irst	-430
+▁spe	-431
+▁care	-432
+��chan	-433
+▁just	-434
+ink	-435
+rie	-436
+spe	-437
+▁As	-438
+ment	-439
+▁say	-440
+▁see	-441
+ember	-442
+▁comp	-443
+▁understand	-444
+▁C	-445
+▁co	-446
+▁us	-447
+▁let	-448
+▁pleas	-449
+▁thine	-450
+▁bet	-451
+▁grie	-452
+▁like	-453
+▁If	-454
+▁sa	-455
+▁des	-456
+▁rem	-457
+ether	-458
+▁body	-459
+ations	-460
+ry	-461
+ue	-462
+▁uni	-463
+▁dost	-464
+ph	-465
+ition	-466
+▁bec	-467
+▁had	-468
+▁way	-469
+▁first	-470
+▁though	-471
+li	-472
+sol	-473
+▁its	-474
+▁pass	-475
+ertain	-476
+▁being	-477
+▁think	-478
+▁L	-479
+ning	-480
+ather	-481
+▁been	-482
+▁against	-483
+ser	-484
+▁mad	-485
+iness	-486
+ution	-487
+▁most	-488
+▁right	-489
+▁universe	-490
+ro	-491
+▁ad	-492
+▁cour	-493
+ual	-494
+vid	-495
+▁sp	-496
+cess	-497
+▁about	-498
+▁conce	-499
+▁others	-500
+▁whether	-501
+un	-502
+old	-503
+urt	-504
+▁done	-505
+▁hurt	-506
+gether	-507
+▁great	-508
+▁might	-509
+▁P	-510
+▁truly	-511
+▁better	-512
+▁contin	-513
+ag	-514
+ea	-515
+ty	-516
+ady	-517
+ans	-518
+ick	-519
+ner	-520
+▁That	-521
+▁const	-522
+▁reasonable	-523
+ep	-524
+not	-525
+ving	-526
+▁actions	-527
+▁certain	-528
+ab	-529
+ig	-530
+▁D	-531
+eed	-532
+▁fo	-533
+▁ph	-534
+▁beg	-535
+▁ear	-536
+▁sin	-537
+▁need	-538
+▁trou	-539
+▁creat	-540
+▁before	-541
+▁object	-542
+▁general	-543
+▁pleasure	-544
+if	-545
+oin	-546
+▁So	-547
+rough	-548
+▁made	-549
+ol	-550
+▁qu	-551
+ound	-552
+▁imp	-553
+ittle	-554
+▁able	-555
+▁take	-556
+▁whom	-557
+▁ordin	-558
+▁cannot	-559
+▁indeed	-560
+▁little	-561
+▁content	-562
+▁through	-563
+▁creature	-564
+▁understanding	-565
+▁int	-566
+spect	-567
+▁come	-568
+▁soon	-569
+▁action	-570
+▁herself	-571
+▁towards	-572
+wh	-573
+▁E	-574
+oss	-575
+ient	-576
+▁est	-577
+▁evil	-578
+▁doest	-579
+▁desire	-580
+ily	-581
+ten	-582
+ways	-583
+▁Gods	-584
+▁requ	-585
+osition	-586
+▁within	-587
+▁opinion	-588
+▁trouble	-589
+ild	-590
+▁car	-591
+▁free	-592
+▁once	-593
+▁cause	-594
+▁concer	-595
+▁'	-596
+ers	-597
+▁To	-598
+▁occ	-599
+▁off	-600
+▁put	-601
+▁find	-602
+▁hind	-603
+▁manner	-604
+cy	-605
+pl	-606
+ces	-607
+lly	-608
+reh	-609
+urn	-610
+ated	-611
+ceed	-612
+▁Thou	-613
+▁mayest	-614
+▁necess	-615
+▁remember	-616
+for	-617
+ilt	-618
+sed	-619
+oint	-620
+ount	-621
+▁cor	-622
+▁This	-623
+▁look	-624
+▁while	-625
+▁always	-626
+▁thought	-627
+▁ordinary	-628
+ace	-629
+ell	-630
+ish	-631
+ple	-632
+ret	-633
+▁How	-634
+▁mere	-635
+▁rest	-636
+ational	-637
+▁course	-638
+wise	-639
+▁pra	-640
+▁speak	-641
+▁subst	-642
+▁subject	-643
+ned	-644
+▁He	-645
+▁el	-646
+▁en	-647
+▁Now	-648
+▁inc	-649
+▁best	-650
+▁disc	-651
+▁dist	-652
+▁here	-653
+▁rather	-654
+▁require	-655
+ilo	-656
+▁fl	-657
+soph	-658
+▁Let	-659
+▁away	-660
+▁over	-661
+▁earth	-662
+▁philo	-663
+▁consist	-664
+▁philosoph	-665
+▁bl	-666
+▁fa	-667
+ense	-668
+pose	-669
+▁ang	-670
+▁pur	-671
+▁bear	-672
+▁prof	-673
+▁soci	-674
+▁occas	-675
+▁still	-676
+▁rational	-677
+▁wo	-678
+▁bel	-679
+▁rep	-680
+▁call	-681
+▁else	-682
+▁fall	-683
+▁pain	-684
+▁wilt	-685
+▁respect	-686
+▁concerning	-687
+▁M	-688
+red	-689
+row	-690
+▁It	-691
+▁adm	-692
+▁order	-693
+cept	-694
+▁equ	-695
+▁conf	-696
+▁said	-697
+▁obser	-698
+▁place	-699
+▁become	-700
+▁hinder	-701
+iv	-702
+ru	-703
+sp	-704
+age	-705
+igh	-706
+ond	-707
+▁ke	-708
+▁beh	-709
+▁gra	-710
+▁hon	-711
+ended	-712
+▁needs	-713
+fferent	-714
+▁change	-715
+▁contem	-716
+▁myself	-717
+▁appoint	-718
+▁together	-719
+▁happiness	-720
+▁substance	-721
+▁disposition	-722
+ne	-723
+oy	-724
+sh	-725
+led	-726
+▁day	-727
+▁fear	-728
+▁word	-729
+▁contr	-730
+▁often	-731
+▁belong	-732
+▁proceed	-733
+▁creatures	-734
+ied	-735
+irt	-736
+ods	-737
+ren	-738
+ates	-739
+▁reg	-740
+▁ret	-741
+▁exam	-742
+▁love	-743
+▁stand	-744
+▁truth	-745
+ossible	-746
+▁suffer	-747
+▁worldly	-748
+▁properly	-749
+▁presently	-750
+ie	-751
+we	-752
+oid	-753
+ood	-754
+ove	-755
+▁Of	-756
+ract	-757
+tern	-758
+▁mod	-759
+▁mot	-760
+▁down	-761
+▁resol	-762
+itution	-763
+▁constitution	-764
+cr	-765
+fer	-766
+ite	-767
+ute	-768
+▁fe	-769
+▁fu	-770
+▁die	-771
+▁set	-772
+▁why	-773
+ffect	-774
+▁keep	-775
+▁worth	-776
+▁appreh	-777
+▁honour	-778
+▁regard	-779
+▁wonder	-780
+▁justice	-781
+▁everything	-782
+by	-783
+of	-784
+ide	-785
+ile	-786
+▁wr	-787
+acul	-788
+form	-789
+▁ser	-790
+ments	-791
+▁indi	-792
+▁sens	-793
+▁suff	-794
+▁thus	-795
+▁virt	-796
+▁work	-797
+itable	-798
+▁child	-799
+▁ready	-800
+▁behold	-801
+▁conceit	-802
+▁perfect	-803
+▁contrary	-804
+▁Whatsoever	-805
+.'	-806
+gn	-807
+▁ac	-808
+▁br	-809
+pend	-810
+ress	-811
+▁mut	-812
+▁form	-813
+▁gods	-814
+▁facul	-815
+▁mayst	-816
+▁sense	-817
+vidence	-818
+▁opinions	-819
+▁thoughts	-820
+ek	-821
+lp	-822
+sc	-823
+ont	-824
+▁In	-825
+▁em	-826
+▁pu	-827
+▁sl	-828
+rupt	-829
+vers	-830
+▁tak	-831
+▁cond	-832
+▁dead	-833
+▁foll	-834
+▁meek	-835
+▁oper	-836
+▁seek	-837
+▁want	-838
+▁begin	-839
+▁canst	-840
+▁sever	-841
+▁worse	-842
+▁effect	-843
+▁please	-844
+▁praise	-845
+bl	-846
+ave	-847
+ial	-848
+imes	-849
+sist	-850
+vour	-851
+▁bre	-852
+▁div	-853
+▁jud	-854
+▁sim	-855
+ready	-856
+▁help	-857
+▁turn	-858
+rought	-859
+▁among	-860
+▁chang	-861
+▁gover	-862
+▁follow	-863
+▁already	-864
+▁because	-865
+▁continu	-866
+▁purpose	-867
+▁occasion	-868
+▁indifferent	-869
+aw	-870
+ec	-871
+eem	-872
+ses	-873
+iend	-874
+ived	-875
+sequ	-876
+▁cut	-877
+▁dep	-878
+▁ele	-879
+icked	-880
+▁alto	-881
+▁seem	-882
+▁whose	-883
+▁estate	-884
+▁esteem	-885
+▁friend	-886
+asonable	-887
+▁several	-888
+▁altogether	-889
+fu	-890
+ip	-891
+ix	-892
+ius	-893
+▁gr	-894
+fend	-895
+less	-896
+over	-897
+uous	-898
+void	-899
+▁aff	-900
+▁cur	-901
+▁fan	-902
+emper	-903
+▁deli	-904
+▁imag	-905
+▁last	-906
+chance	-907
+▁angry	-908
+▁divid	-909
+▁grief	-910
+▁parts	-911
+ination	-912
+▁depend	-913
+▁offend	-914
+▁wicked	-915
+▁brought	-916
+▁observe	-917
+▁affected	-918
+▁appointed	-919
+▁naturally	-920
+▁necessity	-921
+ap	-922
+ety	-923
+rib	-924
+the	-925
+▁Ag	-926
+blic	-927
+ents	-928
+osed	-929
+▁All	-930
+▁exp	-931
+▁ign	-932
+▁law	-933
+▁inst	-934
+▁show	-935
+▁tend	-936
+▁appear	-937
+inations	-938
+▁consequ	-939
+▁whereof	-940
+▁conceive	-941
+▁continue	-942
+▁possible	-943
+act	-944
+ang	-945
+err	-946
+ich	-947
+ory	-948
+▁po	-949
+alth	-950
+ater	-951
+ited	-952
+▁air	-953
+▁bor	-954
+▁dog	-955
+▁ext	-956
+▁sec	-957
+ained	-958
+ceive	-959
+oment	-960
+▁city	-961
+▁incl	-962
+cially	-963
+▁carri	-964
+▁ignor	-965
+▁short	-966
+▁trans	-967
+▁bodies	-968
+▁divine	-969
+▁public	-970
+▁soever	-971
+▁corrupt	-972
+▁deliber	-973
+▁distract	-974
+▁happened	-975
+▁apprehens	-976
+▁necessary	-977
+lf	-978
+ler	-979
+les	-980
+sen	-981
+ult	-982
+▁ey	-983
+ched	-984
+ency	-985
+ides	-986
+oken	-987
+ught	-988
+▁age	-989
+▁eas	-990
+▁fit	-991
+▁new	-992
+▁yea	-993
+▁you	-994
+ities	-995
+itive	-996
+▁char	-997
+▁give	-998
+▁hand	-999
+▁priv	-1000
+▁sort	-1001
+▁Again	-1002
+▁agree	-1003
+▁could	-1004
+▁liber	-1005
+▁souls	-1006
+▁depart	-1007
+▁modest	-1008
+▁moment	-1009
+▁example	-1010
+▁faculty	-1011
+▁whereby	-1012
+▁commonly	-1013
+▁contented	-1014
+▁perchance	-1015
+▁whensoever	-1016
+oc	-1017
+iss	-1018
+▁sc	-1019
+▁sm	-1020
+▁tw	-1021
+ably	-1022
+enet	-1023
+▁exc	-1024
+▁inf	-1025
+▁opp	-1026
+▁pat	-1027
+▁rec	-1028
+▁str	-1029
+ander	-1030
+ately	-1031
+ledge	-1032
+▁espe	-1033
+▁high	-1034
+▁lust	-1035
+▁read	-1036
+▁minds	-1037
+▁since	-1038
+▁sound	-1039
+theless	-1040
+▁extern	-1041
+▁having	-1042
+▁objects	-1043
+▁perform	-1044
+▁likewise	-1045
+▁knowledge	-1046
+▁otherwise	-1047
+▁philosophy	-1048
+ef	-1049
+ash	-1050
+ata	-1051
+min	-1052
+▁Be	-1053
+▁Is	-1054
+▁cl	-1055
+▁ho	-1056
+▁ve	-1057
+ears	-1058
+irit	-1059
+most	-1060
+ough	-1061
+what	-1062
+▁bus	-1063
+▁enj	-1064
+▁err	-1065
+▁fam	-1066
+▁fur	-1067
+▁imm	-1068
+▁rel	-1069
+▁two	-1070
+fully	-1071
+ortal	-1072
+round	-1073
+usion	-1074
+▁bare	-1075
+▁forg	-1076
+▁gone	-1077
+▁plac	-1078
+ceived	-1079
+icient	-1080
+▁alter	-1081
+▁avoid	-1082
+▁bound	-1083
+▁carry	-1084
+▁griev	-1085
+▁happy	-1086
+▁ought	-1087
+▁accept	-1088
+▁breath	-1089
+▁dissol	-1090
+▁ground	-1091
+▁intent	-1092
+▁preser	-1093
+▁motions	-1094
+▁children	-1095
+▁elements	-1096
+▁esteemed	-1097
+▁sociable	-1098
+▁affections	-1099
+▁especially	-1100
+▁nevertheless	-1101
+ach	-1102
+dle	-1103
+par	-1104
+ude	-1105
+cies	-1106
+enef	-1107
+fort	-1108
+arned	-1109
+eness	-1110
+icity	-1111
+ingly	-1112
+itted	-1113
+▁When	-1114
+▁appl	-1115
+▁base	-1116
+▁born	-1117
+▁deed	-1118
+▁exce	-1119
+▁hold	-1120
+▁seen	-1121
+▁sick	-1122
+crates	-1123
+eavour	-1124
+edient	-1125
+▁apply	-1126
+▁benef	-1127
+▁blood	-1128
+▁didst	-1129
+▁shame	-1130
+▁taken	-1131
+▁years	-1132
+▁endure	-1133
+▁spirit	-1134
+▁spoken	-1135
+▁wholly	-1136
+▁further	-1137
+▁profess	-1138
+▁external	-1139
+▁grievous	-1140
+▁affection	-1141
+▁providence	-1142
+su	-1143
+xt	-1144
+ys	-1145
+irc	-1146
+pet	-1147
+uit	-1148
+uty	-1149
+▁ce	-1150
+▁gl	-1151
+alse	-1152
+erci	-1153
+ered	-1154
+seth	-1155
+▁Con	-1156
+▁Not	-1157
+▁bur	-1158
+▁dec	-1159
+▁del	-1160
+▁har	-1161
+▁mak	-1162
+▁run	-1163
+▁van	-1164
+ances	-1165
+arent	-1166
+ative	-1167
+scrib	-1168
+▁circ	-1169
+▁dest	-1170
+▁dogm	-1171
+▁fire	-1172
+▁less	-1173
+▁mine	-1174
+▁pray	-1175
+▁supp	-1176
+ortion	-1177
+▁alone	-1178
+▁anger	-1179
+▁enjoy	-1180
+▁false	-1181
+▁grant	-1182
+▁heart	-1183
+▁pains	-1184
+▁saith	-1185
+▁simpl	-1186
+▁words	-1187
+▁wrong	-1188
+▁causes	-1189
+▁exerci	-1190
+▁govern	-1191
+▁grieve	-1192
+▁mutual	-1193
+▁temper	-1194
+▁comfort	-1195
+▁convers	-1196
+▁equally	-1197
+▁friends	-1198
+▁liberty	-1199
+▁subsist	-1200
+▁suffice	-1201
+▁wherein	-1202
+▁constant	-1203
+▁commended	-1204
+▁endeavour	-1205
+▁simplicity	-1206
+▁apprehension	-1207
+▁imaginations	-1208
+:)	-1209
+up	-1210
+ail	-1211
+ets	-1212
+ian	-1213
+ict	-1214
+ign	-1215
+ins	-1216
+rom	-1217
+ted	-1218
+▁Or	-1219
+▁hu	-1220
+▁ro	-1221
+fter	-1222
+ians	-1223
+just	-1224
+oice	-1225
+resp	-1226
+▁Why	-1227
+▁fut	-1228
+▁hum	-1229
+▁mem	-1230
+▁mis	-1231
+▁pos	-1232
+umber	-1233
+▁Then	-1234
+▁fort	-1235
+▁full	-1236
+▁sure	-1237
+▁vile	-1238
+▁wret	-1239
+icious	-1240
+itions	-1241
+itious	-1242
+solute	-1243
+vation	-1244
+▁There	-1245
+▁These	-1246
+▁Those	-1247
+▁comes	-1248
+▁compl	-1249
+▁error	-1250
+▁infin	-1251
+▁light	-1252
+▁lives	-1253
+▁pract	-1254
+▁refer	-1255
+▁separ	-1256
+emember	-1257
+▁almost	-1258
+▁disper	-1259
+▁easily	-1260
+▁excell	-1261
+▁future	-1262
+▁return	-1263
+▁virtue	-1264
+▁belongs	-1265
+▁carried	-1266
+▁hurtful	-1267
+▁matters	-1268
+▁outward	-1269
+▁patient	-1270
+▁pleased	-1271
+▁absolute	-1272
+▁sensible	-1273
+▁certainly	-1274
+▁operation	-1275
+▁pleasures	-1276
+▁wickedness	-1277
+,)	-1278
+uc	-1279
+iar	-1280
+ipe	-1281
+ird	-1282
+mon	-1283
+oms	-1284
+ues	-1285
+ven	-1286
+▁Do	-1287
+▁da	-1288
+▁mo	-1289
+▁ru	-1290
+anim	-1291
+ines	-1292
+ings	-1293
+lect	-1294
+osec	-1295
+sess	-1296
+sion	-1297
+ties	-1298
+used	-1299
+▁Pro	-1300
+▁bad	-1301
+▁cle	-1302
+▁old	-1303
+▁red	-1304
+▁rid	-1305
+▁too	-1306
+cient	-1307
+enetr	-1308
+ience	-1309
+ondly	-1310
+ories	-1311
+otten	-1312
+sides	-1313
+where	-1314
+which	-1315
+▁From	-1316
+▁cast	-1317
+▁eyes	-1318
+▁hear	-1319
+▁judg	-1320
+▁mean	-1321
+▁succ	-1322
+▁unre	-1323
+▁above	-1324
+▁cease	-1325
+▁count	-1326
+▁doing	-1327
+▁found	-1328
+▁fruit	-1329
+▁lived	-1330
+lerable	-1331
+ountain	-1332
+▁afford	-1333
+▁commit	-1334
+▁intend	-1335
+▁justly	-1336
+▁living	-1337
+▁motion	-1338
+▁penetr	-1339
+▁places	-1340
+▁prosec	-1341
+▁ancient	-1342
+▁compass	-1343
+▁delight	-1344
+▁divided	-1345
+▁dogmata	-1346
+▁fancies	-1347
+▁passion	-1348
+▁possess	-1349
+▁society	-1350
+▁Socrates	-1351
+▁complain	-1352
+▁contempl	-1353
+▁exercise	-1354
+▁offended	-1355
+▁somewhat	-1356
+▁beginning	-1357
+▁ignorance	-1358
+▁occasions	-1359
+▁profitable	-1360
+▁sufficient	-1361
+),	-1362
+he	-1363
+ts	-1364
+yr	-1365
+bra	-1366
+eer	-1367
+ife	-1368
+raw	-1369
+▁Ep	-1370
+▁Ne	-1371
+▁Se	-1372
+▁sk	-1373
+ancy	-1374
+ical	-1375
+pect	-1376
+ysic	-1377
+▁ash	-1378
+▁due	-1379
+▁fre	-1380
+▁lab	-1381
+▁ost	-1382
+▁sat	-1383
+▁sol	-1384
+▁suc	-1385
+▁ter	-1386
+areth	-1387
+gress	-1388
+ility	-1389
+▁Doth	-1390
+▁beli	-1391
+▁fame	-1392
+▁mast	-1393
+▁past	-1394
+▁prop	-1395
+▁stra	-1396
+▁stud	-1397
+▁wise	-1398
+petual	-1399
+retion	-1400
+▁cheer	-1401
+▁daily	-1402
+▁etern	-1403
+▁means	-1404
+▁seest	-1405
+▁three	-1406
+▁times	-1407
+ousness	-1408
+▁circum	-1409
+▁freely	-1410
+▁inward	-1411
+▁longer	-1412
+▁loving	-1413
+▁member	-1414
+▁mortal	-1415
+▁ostent	-1416
+▁physic	-1417
+▁besides	-1418
+▁changes	-1419
+▁desires	-1420
+▁grieved	-1421
+▁happens	-1422
+▁learned	-1423
+▁propose	-1424
+▁wouldst	-1425
+▁apparent	-1426
+▁disposed	-1427
+▁hindered	-1428
+▁received	-1429
+▁wretched	-1430
+▁forgotten	-1431
+▁generally	-1432
+▁whosoever	-1433
+▁corruption	-1434
+▁ostentation	-1435
+▁unreasonable	-1436
+gg	-1437
+ms	-1438
+sw	-1439
+bly	-1440
+ert	-1441
+ier	-1442
+iol	-1443
+ipp	-1444
+isf	-1445
+iti	-1446
+lst	-1447
+ors	-1448
+▁No	-1449
+▁bo	-1450
+anqu	-1451
+anus	-1452
+bour	-1453
+each	-1454
+inct	-1455
+ired	-1456
+▁Him	-1457
+▁Sec	-1458
+▁add	-1459
+▁dre	-1460
+▁far	-1461
+▁fig	-1462
+▁fly	-1463
+▁ill	-1464
+▁neg	-1465
+▁tyr	-1466
+ently	-1467
+inist	-1468
+ordin	-1469
+orrow	-1470
+▁answ	-1471
+▁came	-1472
+▁cred	-1473
+▁diff	-1474
+▁fell	-1475
+▁fool	-1476
+▁grow	-1477
+▁heed	-1478
+▁main	-1479
+▁prev	-1480
+▁prin	-1481
+illity	-1482
+iously	-1483
+resent	-1484
+tained	-1485
+▁beast	-1486
+▁belie	-1487
+▁bring	-1488
+▁grave	-1489
+▁human	-1490
+▁inter	-1491
+▁leave	-1492
+▁moder	-1493
+▁neigh	-1494
+▁quest	-1495
+▁speed	-1496
+▁teach	-1497
+respond	-1498
+▁buried	-1499
+▁concur	-1500
+▁consum	-1501
+▁destin	-1502
+▁expect	-1503
+▁satisf	-1504
+▁tranqu	-1505
+▁unjust	-1506
+▁whilst	-1507
+▁changed	-1508
+▁commend	-1509
+▁contemn	-1510
+▁portion	-1511
+▁success	-1512
+▁applause	-1513
+▁converse	-1514
+▁judgment	-1515
+▁practice	-1516
+▁relation	-1517
+▁straight	-1518
+▁excellent	-1519
+▁expedient	-1520
+▁perpetual	-1521
+▁willingly	-1522
+▁Providence	-1523
+▁consequent	-1524
+▁correspond	-1525
+▁discretion	-1526
+▁deliberation	-1527
+▁tranquillity	-1528
+▁consideration	-1529
+og	-1530
+ym	-1531
+ank	-1532
+ari	-1533
+ato	-1534
+den	-1535
+edy	-1536
+ern	-1537
+omp	-1538
+ron	-1539
+une	-1540
+▁Ch	-1541
+▁Pl	-1542
+▁sw	-1543
+aked	-1544
+ants	-1545
+ased	-1546
+eter	-1547
+ield	-1548
+insm	-1549
+ixed	-1550
+leep	-1551
+owed	-1552
+rect	-1553
+ship	-1554
+▁His	-1555
+▁One	-1556
+▁few	-1557
+▁fle	-1558
+▁mag	-1559
+▁mer	-1560
+▁mor	-1561
+▁pas	-1562
+▁swe	-1563
+▁vol	-1564
+ident	-1565
+itude	-1566
+▁Such	-1567
+▁dust	-1568
+▁fore	-1569
+▁hard	-1570
+▁lose	-1571
+▁near	-1572
+▁rash	-1573
+▁seed	-1574
+▁self	-1575
+▁sins	-1576
+▁susp	-1577
+▁vain	-1578
+▁viol	-1579
+▁wild	-1580
+▁wont	-1581
+leased	-1582
+rained	-1583
+▁appet	-1584
+▁beget	-1585
+▁court	-1586
+▁embra	-1587
+▁fancy	-1588
+▁hindr	-1589
+▁judge	-1590
+▁kinsm	-1591
+▁known	-1592
+▁lieth	-1593
+▁mater	-1594
+▁sleep	-1595
+▁state	-1596
+▁until	-1597
+itherto	-1598
+▁Either	-1599
+▁called	-1600
+▁credit	-1601
+▁direct	-1602
+▁except	-1603
+▁former	-1604
+▁labour	-1605
+▁liveth	-1606
+▁meekly	-1607
+▁number	-1608
+▁secret	-1609
+▁senses	-1610
+▁simple	-1611
+▁vanity	-1612
+▁account	-1613
+▁benefit	-1614
+▁chances	-1615
+▁compreh	-1616
+▁curious	-1617
+▁however	-1618
+▁modesty	-1619
+▁neglect	-1620
+▁private	-1621
+▁succeed	-1622
+▁tending	-1623
+▁cheerful	-1624
+▁contempt	-1625
+▁eternity	-1626
+▁fountain	-1627
+▁governed	-1628
+▁hitherto	-1629
+▁infinite	-1630
+▁material	-1631
+▁meekness	-1632
+▁perceive	-1633
+▁pleasing	-1634
+▁question	-1635
+▁confusion	-1636
+▁neighbour	-1637
+▁reference	-1638
+▁represent	-1639
+▁sensitive	-1640
+▁absolutely	-1641
+▁displeased	-1642
+▁impossible	-1643
+▁possession	-1644
+▁transgress	-1645
+▁dissolution	-1646
+▁philosophers	-1647
+,'	-1648
+ax	-1649
+el	-1650
+po	-1651
+aud	-1652
+ief	-1653
+lig	-1654
+ock	-1655
+uce	-1656
+use	-1657
+▁Ad	-1658
+▁gu	-1659
+▁te	-1660
+▁va	-1661
+amed	-1662
+ames	-1663
+ared	-1664
+aten	-1665
+ctif	-1666
+draw	-1667
+edim	-1668
+ingl	-1669
+inus	-1670
+itus	-1671
+nect	-1672
+nest	-1673
+sens	-1674
+thly	-1675
+vert	-1676
+vity	-1677
+▁Ant	-1678
+▁amb	-1679
+▁bea	-1680
+▁cal	-1681
+▁ent	-1682
+▁get	-1683
+▁gre	-1684
+▁inv	-1685
+▁joy	-1686
+▁sym	-1687
+▁vis	-1688
+anger	-1689
+ences	-1690
+ished	-1691
+itten	-1692
+otted	-1693
+ously	-1694
+sider	-1695
+ually	-1696
+unity	-1697
+▁They	-1698
+▁Thus	-1699
+▁citi	-1700
+▁excl	-1701
+▁fair	-1702
+▁gent	-1703
+▁glad	-1704
+▁hour	-1705
+▁lead	-1706
+▁magn	-1707
+▁next	-1708
+▁peri	-1709
+▁qual	-1710
+▁serv	-1711
+▁stir	-1712
+▁stri	-1713
+▁subm	-1714
+▁ways	-1715
+apable	-1716
+father	-1717
+itting	-1718
+ontent	-1719
+orious	-1720
+ounded	-1721
+▁Where	-1722
+▁atoms	-1723
+▁deter	-1724
+▁famil	-1725
+▁fates	-1726
+▁flesh	-1727
+▁forth	-1728
+▁fully	-1729
+▁least	-1730
+▁makes	-1731
+▁names	-1732
+▁seems	-1733
+▁study	-1734
+▁takes	-1735
+▁usual	-1736
+▁wills	-1737
+animity	-1738
+nection	-1739
+scribed	-1740
+▁differ	-1741
+▁dispos	-1742
+▁fellow	-1743
+▁kindly	-1744
+▁mother	-1745
+▁partak	-1746
+▁rectif	-1747
+▁remain	-1748
+▁righte	-1749
+▁vaingl	-1750
+▁Neither	-1751
+▁command	-1752
+▁discern	-1753
+▁earnest	-1754
+▁earthly	-1755
+▁examine	-1756
+▁impedim	-1757
+▁natures	-1758
+▁receive	-1759
+▁rightly	-1760
+▁suppose	-1761
+▁thereby	-1762
+▁upwards	-1763
+▁Consider	-1764
+▁administ	-1765
+▁commonwe	-1766
+▁resolved	-1767
+▁shameful	-1768
+▁troubles	-1769
+▁virtuous	-1770
+▁withdraw	-1771
+▁agreeable	-1772
+▁carefully	-1773
+▁different	-1774
+▁readiness	-1775
+▁satisfied	-1776
+▁shouldest	-1777
+▁alteration	-1778
+▁connection	-1779
+▁perfection	-1780
+▁profession	-1781
+▁resolution	-1782
+▁appointment	-1783
+▁continually	-1784
+▁philosopher	-1785
+▁righteousness	-1786
+▁understandings	-1787
+..	-1788
+av	-1789
+ack	-1790
+ade	-1791
+ars	-1792
+cer	-1793
+cip	-1794
+gth	-1795
+ift	-1796
+lex	-1797
+mur	-1798
+oud	-1799
+pes	-1800
+pri	-1801
+rum	-1802
+tle	-1803
+xed	-1804
+▁Ph	-1805
+▁cr	-1806
+▁dr	-1807
+▁ir	-1808
+ator	-1809
+aves	-1810
+bear	-1811
+ches	-1812
+ibly	-1813
+ices	-1814
+imus	-1815
+path	-1816
+rain	-1817
+sand	-1818
+uted	-1819
+with	-1820
+▁Her	-1821
+▁eff	-1822
+▁exh	-1823
+▁ing	-1824
+▁led	-1825
+▁lim	-1826
+▁mur	-1827
+▁oft	-1828
+▁rej	-1829
+▁sac	-1830
+▁spr	-1831
+▁ste	-1832
+▁sum	-1833
+▁sun	-1834
+▁sur	-1835
+▁tre	-1836
+▁una	-1837
+▁unc	-1838
+▁wom	-1839
+after	-1840
+ature	-1841
+ement	-1842
+eries	-1843
+iddle	-1844
+itate	-1845
+phron	-1846
+sence	-1847
+vided	-1848
+▁Alex	-1849
+▁More	-1850
+▁cure	-1851
+▁days	-1852
+▁open	-1853
+▁pure	-1854
+▁ripe	-1855
+▁rule	-1856
+▁tree	-1857
+▁unse	-1858
+▁void	-1859
+▁volu	-1860
+▁wind	-1861
+ention	-1862
+idents	-1863
+oninus	-1864
+though	-1865
+▁Plato	-1866
+▁admit	-1867
+▁begun	-1868
+▁clear	-1869
+▁concl	-1870
+▁equal	-1871
+▁fault	-1872
+▁hadst	-1873
+▁loose	-1874
+▁noise	-1875
+▁persu	-1876
+▁plain	-1877
+▁pride	-1878
+▁proud	-1879
+▁quick	-1880
+▁reser	-1881
+▁serve	-1882
+▁sight	-1883
+▁solid	-1884
+▁third	-1885
+▁water	-1886
+iration	-1887
+▁attain	-1888
+▁averse	-1889
+▁beasts	-1890
+▁beauty	-1891
+▁condem	-1892
+▁deaths	-1893
+▁father	-1894
+▁gentle	-1895
+▁health	-1896
+▁herein	-1897
+▁indign	-1898
+▁murmur	-1899
+▁resist	-1900
+▁sooner	-1901
+▁stands	-1902
+▁strong	-1903
+▁united	-1904
+▁wander	-1905
+▁whence	-1906
+sensible	-1907
+▁Whether	-1908
+▁believe	-1909
+▁concern	-1910
+▁contain	-1911
+▁embrace	-1912
+▁flatter	-1913
+▁granted	-1914
+▁gravity	-1915
+▁instrum	-1916
+▁orderly	-1917
+▁resolve	-1918
+▁suspect	-1919
+▁whither	-1920
+▁Moreover	-1921
+▁goodness	-1922
+▁ignorant	-1923
+▁mightest	-1924
+▁observed	-1925
+▁presents	-1926
+▁provided	-1927
+▁subjects	-1928
+▁thousand	-1929
+▁Alexander	-1930
+▁Antoninus	-1931
+▁accidents	-1932
+▁continual	-1933
+▁faculties	-1934
+▁intention	-1935
+▁penetrate	-1936
+▁requireth	-1937
+▁afterwards	-1938
+▁dispersion	-1939
+▁operations	-1940
+▁seasonable	-1941
+▁consequence	-1942
+▁indignation	-1943
+▁intolerable	-1944
+▁conversation	-1945
+▁dispositions	-1946
+▁distractions	-1947
+be	-1948
+ha	-1949
+io	-1950
+ng	-1951
+sa	-1952
+te	-1953
+ux	-1954
+any	-1955
+ati	-1956
+bit	-1957
+bod	-1958
+ced	-1959
+ery	-1960
+gar	-1961
+iet	-1962
+ked	-1963
+nto	-1964
+rif	-1965
+rop	-1966
+sem	-1967
+▁An	-1968
+▁By	-1969
+▁bu	-1970
+▁dy	-1971
+▁ut	-1972
+aths	-1973
+augh	-1974
+body	-1975
+dden	-1976
+etus	-1977
+happ	-1978
+iled	-1979
+ilig	-1980
+inal	-1981
+ined	-1982
+ones	-1983
+oper	-1984
+ruct	-1985
+ssed	-1986
+uity	-1987
+ween	-1988
+▁Ath	-1989
+▁bed	-1990
+▁dur	-1991
+▁hor	-1992
+▁lot	-1993
+▁mar	-1994
+▁med	-1995
+▁ref	-1996
+▁rul	-1997
+▁smo	-1998
+▁tem	-1999
+▁ten	-2000
+▁vul	-2001
+arily	-2002
+cious	-2003
+icial	-2004
+lfare	-2005
+udent	-2006
+▁acqu	-2007
+▁ages	-2008
+▁case	-2009
+▁ease	-2010
+▁ends	-2011
+▁fash	-2012
+▁flow	-2013
+▁food	-2014
+▁hate	-2015
+▁hunt	-2016
+▁idle	-2017
+▁join	-2018
+▁meat	-2019
+▁meth	-2020
+▁none	-2021
+▁obsc	-2022
+▁obst	-2023
+▁play	-2024
+▁poet	-2025
+▁sake	-2026
+▁send	-2027
+▁spee	-2028
+▁subt	-2029
+▁tell	-2030
+▁vine	-2031
+▁wife	-2032
+action	-2033
+enance	-2034
+ically	-2035
+leness	-2036
+mation	-2037
+ration	-2038
+rength	-2039
+▁After	-2040
+▁First	-2041
+▁blame	-2042
+▁dieth	-2043
+▁dilig	-2044
+▁ended	-2045
+▁exact	-2046
+▁excre	-2047
+▁exper	-2048
+▁flock	-2049
+▁force	-2050
+▁fresh	-2051
+▁hands	-2052
+▁heard	-2053
+▁ingen	-2054
+▁knows	-2055
+▁lusts	-2056
+▁naked	-2057
+▁point	-2058
+▁repro	-2059
+▁seeth	-2060
+▁small	-2061
+▁super	-2062
+▁throw	-2063
+▁unity	-2064
+▁young	-2065
+▁youth	-2066
+entimes	-2067
+inesses	-2068
+▁accuse	-2069
+▁answer	-2070
+▁compas	-2071
+▁corner	-2072
+▁equity	-2073
+▁events	-2074
+▁favour	-2075
+▁grapes	-2076
+▁harmon	-2077
+▁letter	-2078
+▁making	-2079
+▁merely	-2080
+▁middle	-2081
+▁prefer	-2082
+▁report	-2083
+▁series	-2084
+▁sorrow	-2085
+▁sudden	-2086
+▁unhapp	-2087
+▁vulgar	-2088
+▁wealth	-2089
+▁withal	-2090
+▁worthy	-2091
+▁ashamed	-2092
+▁between	-2093
+▁breathe	-2094
+▁capable	-2095
+▁country	-2096
+▁distrib	-2097
+▁fashion	-2098
+▁feareth	-2099
+▁fitting	-2100
+▁fortune	-2101
+▁greater	-2102
+▁impious	-2103
+▁incline	-2104
+▁kindred	-2105
+▁members	-2106
+▁mention	-2107
+▁rectify	-2108
+▁rejoice	-2109
+▁sympath	-2110
+▁welfare	-2111
+▁Secondly	-2112
+▁allotted	-2113
+▁although	-2114
+▁business	-2115
+▁contract	-2116
+▁familiar	-2117
+▁intended	-2118
+▁ordained	-2119
+▁powerful	-2120
+▁proceeds	-2121
+▁secondly	-2122
+▁shouldst	-2123
+▁speaking	-2124
+▁strength	-2125
+▁ambitious	-2126
+▁committed	-2127
+▁dispersed	-2128
+▁dissolved	-2129
+▁distemper	-2130
+▁hereafter	-2131
+▁patiently	-2132
+▁physician	-2133
+▁preserved	-2134
+▁vainglory	-2135
+▁appearance	-2136
+▁businesses	-2137
+▁charitable	-2138
+▁consisting	-2139
+▁discontent	-2140
+▁everywhere	-2141
+▁experience	-2142
+▁government	-2143
+▁intentions	-2144
+▁oftentimes	-2145
+▁prescribed	-2146
+▁substances	-2147
+▁affectation	-2148
+▁continuance	-2149
+▁countenance	-2150
+▁inclination	-2151
+▁reservation	-2152
+▁commonwealth	-2153
+▁contemplation	-2154
+II	-2155
+eo	-2156
+ex	-2157
+op	-2158
+acl	-2159
+aim	-2160
+air	-2161
+ams	-2162
+ann	-2163
+edi	-2164
+eso	-2165
+ged	-2166
+ics	-2167
+ior	-2168
+iot	-2169
+isc	-2170
+isd	-2171
+ixt	-2172
+med	-2173
+omm	-2174
+por	-2175
+rel	-2176
+rem	-2177
+rob	-2178
+say	-2179
+son	-2180
+sum	-2181
+thy	-2182
+uct	-2183
+uth	-2184
+wor	-2185
+ypo	-2186
+▁aw	-2187
+▁es	-2188
+▁gi	-2189
+▁om	-2190
+▁rh	-2191
+acle	-2192
+arch	-2193
+bled	-2194
+cere	-2195
+ecul	-2196
+ened	-2197
+enes	-2198
+here	-2199
+illy	-2200
+ploy	-2201
+reet	-2202
+rest	-2203
+uced	-2204
+ular	-2205
+will	-2206
+▁Can	-2207
+▁Div	-2208
+▁See	-2209
+▁Thy	-2210
+▁ago	-2211
+▁att	-2212
+▁col	-2213
+▁dri	-2214
+▁enc	-2215
+▁env	-2216
+▁eye	-2217
+▁fix	-2218
+▁fug	-2219
+▁lie	-2220
+▁pie	-2221
+▁son	-2222
+▁veh	-2223
+▁war	-2224
+aketh	-2225
+allen	-2226
+asure	-2227
+aught	-2228
+cipal	-2229
+ereso	-2230
+erves	-2231
+forth	-2232
+icion	-2233
+ierce	-2234
+ippus	-2235
+iring	-2236
+isure	-2237
+ndred	-2238
+osity	-2239
+oward	-2240
+stood	-2241
+thank	-2242
+uthor	-2243
+ypocr	-2244
+▁Adri	-2245
+▁Hast	-2246
+▁Well	-2247
+▁Will	-2248
+▁alas	-2249
+▁arts	-2250
+▁bran	-2251
+▁deal	-2252
+▁dear	-2253
+▁decl	-2254
+▁duty	-2255
+▁easy	-2256
+▁fors	-2257
+▁gain	-2258
+▁hope	-2259
+▁lest	-2260
+▁move	-2261
+▁name	-2262
+▁ones	-2263
+▁orig	-2264
+▁pomp	-2265
+▁quar	-2266
+▁reco	-2267
+▁refu	-2268
+▁safe	-2269
+▁side	-2270
+▁sign	-2271
+▁thor	-2272
+▁town	-2273
+▁unch	-2274
+▁used	-2275
+▁wert	-2276
+etimes	-2277
+ission	-2278
+ptuous	-2279
+scribe	-2280
+ulsion	-2281
+▁Epict	-2282
+▁admon	-2283
+▁ashes	-2284
+▁beest	-2285
+▁cares	-2286
+▁chief	-2287
+▁close	-2288
+▁hence	-2289
+▁horse	-2290
+▁mixed	-2291
+▁moved	-2292
+▁pecul	-2293
+▁quiet	-2294
+▁quite	-2295
+▁relig	-2296
+▁sheep	-2297
+▁skill	-2298
+▁smoke	-2299
+▁stage	-2300
+▁sweet	-2301
+▁tract	-2302
+▁trees	-2303
+▁wrest	-2304
+▁yield	-2305
+aclitus	-2306
+idental	-2307
+operate	-2308
+reation	-2309
+▁Divine	-2310
+▁Nature	-2311
+▁barely	-2312
+▁danger	-2313
+▁desert	-2314
+▁dissem	-2315
+▁divers	-2316
+▁divide	-2317
+▁employ	-2318
+▁famous	-2319
+▁flocks	-2320
+▁hypocr	-2321
+▁immedi	-2322
+▁intell	-2323
+▁master	-2324
+▁memory	-2325
+▁method	-2326
+▁opport	-2327
+▁period	-2328
+▁person	-2329
+▁rashly	-2330
+▁reduce	-2331
+▁speech	-2332
+▁spring	-2333
+▁submit	-2334
+▁surely	-2335
+▁surpri	-2336
+▁thence	-2337
+▁tyrann	-2338
+▁useful	-2339
+▁vanish	-2340
+ableness	-2341
+standing	-2342
+thankful	-2343
+voidable	-2344
+▁However	-2345
+▁acquain	-2346
+▁careful	-2347
+▁depends	-2348
+▁extinct	-2349
+▁forbear	-2350
+▁hundred	-2351
+▁immense	-2352
+▁kinsmen	-2353
+▁leisure	-2354
+▁masters	-2355
+▁measure	-2356
+▁morning	-2357
+▁nations	-2358
+▁notwith	-2359
+▁prevent	-2360
+▁privile	-2361
+▁uniform	-2362
+▁virtues	-2363
+▁whereas	-2364
+▁whilest	-2365
+▁written	-2366
+▁wronged	-2367
+eresoever	-2368
+▁Adrianus	-2369
+▁becometh	-2370
+▁deceived	-2371
+▁discreet	-2372
+▁duration	-2373
+▁fugitive	-2374
+▁immortal	-2375
+▁instruct	-2376
+▁memories	-2377
+▁moreover	-2378
+▁obtained	-2379
+▁passions	-2380
+▁peculiar	-2381
+▁pleasant	-2382
+▁preserve	-2383
+▁separate	-2384
+▁sickness	-2385
+▁sympathy	-2386
+▁troubled	-2387
+▁wouldest	-2388
+▁Epictetus	-2389
+▁conceived	-2390
+▁condition	-2391
+▁downwards	-2392
+▁hindrance	-2393
+▁mentioned	-2394
+▁principal	-2395
+▁separated	-2396
+▁suspicion	-2397
+▁tolerable	-2398
+▁violently	-2399
+▁Heraclitus	-2400
+▁compasseth	-2401
+▁comprehend	-2402
+▁conversion	-2403
+▁destinated	-2404
+▁difference	-2405
+▁generation	-2406
+▁henceforth	-2407
+▁impediment	-2408
+▁instrument	-2409
+▁ordinarily	-2410
+▁restrained	-2411
+▁understood	-2412
+▁unthankful	-2413
+▁voluptuous	-2414
+▁alterations	-2415
+▁deliberated	-2416
+▁immediately	-2417
+▁magnanimity	-2418
+▁prosecution	-2419
+▁unavoidable	-2420
+▁administered	-2421
+▁consummation	-2422
+▁discontented	-2423
+▁inclinations	-2424
+▁particularly	-2425
+▁apprehensions	-2426
+▁correspondence	-2427
+▁correspondency	-2428
+▁notwithstanding	-2429
+It	-2430
+bt	-2431
+ey	-2432
+ft	-2433
+ox	-2434
+ug	-2435
+ait	-2436
+ald	-2437
+alk	-2438
+arp	-2439
+eep	-2440
+fet	-2441
+her	-2442
+jud	-2443
+ken	-2444
+nal	-2445
+nel	-2446
+nes	-2447
+new	-2448
+ric	-2449
+rid	-2450
+sur	-2451
+▁At	-2452
+▁De	-2453
+▁ap	-2454
+▁du	-2455
+▁id	-2456
+ails	-2457
+alls	-2458
+ange	-2459
+angu	-2460
+arce	-2461
+cend	-2462
+etor	-2463
+ffed	-2464
+icur	-2465
+icus	-2466
+iety	-2467
+ific	-2468
+ifts	-2469
+iful	-2470
+ists	-2471
+iven	-2472
+lest	-2473
+lete	-2474
+like	-2475
+ompe	-2476
+oose	-2477
+ouse	-2478
+raid	-2479
+rawn	-2480
+rent	-2481
+road	-2482
+same	-2483
+surd	-2484
+tled	-2485
+yond	-2486
+▁Aug	-2487
+▁Dem	-2488
+▁Max	-2489
+▁Out	-2490
+▁adv	-2491
+▁arm	-2492
+▁ask	-2493
+▁ass	-2494
+▁bee	-2495
+▁bra	-2496
+▁cas	-2497
+▁cha	-2498
+▁dam	-2499
+▁dut	-2500
+▁eat	-2501
+▁fro	-2502
+▁god	-2503
+▁hol	-2504
+▁inh	-2505
+▁inn	-2506
+▁mal	-2507
+▁peo	-2508
+▁pop	-2509
+▁sco	-2510
+▁sea	-2511
+▁som	-2512
+▁tor	-2513
+▁tum	-2514
+▁vex	-2515
+▁vig	-2516
+▁wat	-2517
+aries	-2518
+arrow	-2519
+arted	-2520
+atioc	-2521
+avish	-2522
+cheth	-2523
+ggest	-2524
+ifest	-2525
+ifold	-2526
+ingle	-2527
+isite	-2528
+mined	-2529
+onour	-2530
+orems	-2531
+ourse	-2532
+sumed	-2533
+thers	-2534
+upisc	-2535
+urity	-2536
+ustus	-2537
+▁Cont	-2538
+▁Dost	-2539
+▁Ever	-2540
+▁Here	-2541
+▁Pius	-2542
+▁Take	-2543
+▁Wilt	-2544
+▁With	-2545
+▁acts	-2546
+▁anim	-2547
+▁back	-2548
+▁book	-2549
+▁busy	-2550
+▁calm	-2551
+▁cess	-2552
+▁curi	-2553
+▁cust	-2554
+▁desp	-2555
+▁does	-2556
+▁drop	-2557
+▁fain	-2558
+▁fast	-2559
+▁feet	-2560
+▁figs	-2561
+▁flux	-2562
+▁gave	-2563
+▁hang	-2564
+▁head	-2565
+▁issu	-2566
+▁kept	-2567
+▁loss	-2568
+▁mass	-2569
+▁meet	-2570
+▁mild	-2571
+▁mist	-2572
+▁mult	-2573
+▁myst	-2574
+▁obey	-2575
+▁prud	-2576
+▁rain	-2577
+▁reap	-2578
+▁resp	-2579
+▁room	-2580
+▁save	-2581
+▁ship	-2582
+▁shut	-2583
+▁stay	-2584
+▁till	-2585
+▁tong	-2586
+▁took	-2587
+▁unle	-2588
+▁unne	-2589
+▁vice	-2590
+antime	-2591
+ension	-2592
+erable	-2593
+ferior	-2594
+fetime	-2595
+icurus	-2596
+nesses	-2597
+oughly	-2598
+rupted	-2599
+utions	-2600
+worthy	-2601
+▁Pompe	-2602
+▁Which	-2603
+▁abide	-2604
+▁amiss	-2605
+▁apost	-2606
+▁appli	-2607
+▁appro	-2608
+▁borne	-2609
+▁bount	-2610
+▁calam	-2611
+▁coher	-2612
+▁cover	-2613
+▁cured	-2614
+▁drawn	-2615
+▁exhal	-2616
+▁extra	-2617
+▁fable	-2618
+▁faith	-2619
+▁falls	-2620
+▁flood	-2621
+▁gifts	-2622
+▁given	-2623
+▁harsh	-2624
+▁imper	-2625
+▁langu	-2626
+▁laugh	-2627
+▁moist	-2628
+▁moral	-2629
+▁occup	-2630
+▁piece	-2631
+▁renew	-2632
+▁round	-2633
+▁rules	-2634
+▁sharp	-2635
+▁shows	-2636
+▁silly	-2637
+▁stick	-2638
+▁stret	-2639
+▁teeth	-2640
+▁tenet	-2641
+▁thank	-2642
+▁towns	-2643
+▁walls	-2644
+▁weary	-2645
+▁write	-2646
+ability	-2647
+astness	-2648
+cessant	-2649
+cessary	-2650
+fathers	-2651
+iculous	-2652
+istence	-2653
+itation	-2654
+ulation	-2655
+utation	-2656
+▁Others	-2657
+▁Though	-2658
+▁abroad	-2659
+▁absurd	-2660
+▁access	-2661
+▁afraid	-2662
+▁assent	-2663
+▁author	-2664
+▁basely	-2665
+▁bestow	-2666
+▁beyond	-2667
+▁bounds	-2668
+▁chance	-2669
+▁choose	-2670
+▁colour	-2671
+▁comely	-2672
+▁damage	-2673
+▁divert	-2674
+▁dreams	-2675
+▁duties	-2676
+▁enough	-2677
+▁excess	-2678
+▁forget	-2679
+▁formal	-2680
+▁higher	-2681
+▁honest	-2682
+▁joined	-2683
+▁lastly	-2684
+▁latter	-2685
+▁leaves	-2686
+▁misery	-2687
+▁nerves	-2688
+▁notice	-2689
+▁overse	-2690
+▁partly	-2691
+▁passed	-2692
+▁people	-2693
+▁plants	-2694
+▁prejud	-2695
+▁profit	-2696
+▁ratioc	-2697
+▁refuge	-2698
+▁repent	-2699
+▁retire	-2700
+▁scarce	-2701
+▁season	-2702
+▁single	-2703
+▁stones	-2704
+▁strive	-2705
+▁tender	-2706
+▁tumult	-2707
+▁unchar	-2708
+▁weight	-2709
+▁Another	-2710
+▁Maximus	-2711
+▁ability	-2712
+▁anyways	-2713
+▁applaud	-2714
+▁beareth	-2715
+▁charity	-2716
+▁clearly	-2717
+▁consent	-2718
+▁courage	-2719
+▁foolish	-2720
+▁forsake	-2721
+▁harmony	-2722
+▁highest	-2723
+▁infinit	-2724
+▁instant	-2725
+▁kinsman	-2726
+▁knowest	-2727
+▁knoweth	-2728
+▁liberal	-2729
+▁limited	-2730
+▁meaning	-2731
+▁obscure	-2732
+▁ordered	-2733
+▁passeth	-2734
+▁popular	-2735
+▁praised	-2736
+▁praises	-2737
+▁quarrel	-2738
+▁quickly	-2739
+▁resumed	-2740
+▁retired	-2741
+▁seeking	-2742
+▁sensual	-2743
+▁sincere	-2744
+▁slavish	-2745
+▁strange	-2746
+▁subtile	-2747
+▁suggest	-2748
+▁unhappy	-2749
+▁vicious	-2750
+▁visible	-2751
+▁wonders	-2752
+atenation	-2753
+▁Augustus	-2754
+▁Epicurus	-2755
+▁answered	-2756
+▁anywhere	-2757
+▁carriage	-2758
+▁creation	-2759
+▁diffused	-2760
+▁effected	-2761
+▁gracious	-2762
+▁greatest	-2763
+▁holiness	-2764
+▁inferior	-2765
+▁inwardly	-2766
+▁language	-2767
+▁lifetime	-2768
+▁maintain	-2769
+▁manifest	-2770
+▁manifold	-2771
+▁meantime	-2772
+▁mildness	-2773
+▁mutation	-2774
+▁obstacle	-2775
+▁original	-2776
+▁patience	-2777
+▁property	-2778
+▁purposes	-2779
+▁reported	-2780
+▁reproach	-2781
+▁required	-2782
+▁restrain	-2783
+▁sociably	-2784
+▁speaketh	-2785
+▁standing	-2786
+▁subordin	-2787
+▁theorems	-2788
+▁vehement	-2789
+▁wrestler	-2790
+whatsoever	-2791
+▁accepting	-2792
+▁apprehend	-2793
+▁coherence	-2794
+▁community	-2795
+▁concupisc	-2796
+▁condemned	-2797
+▁corrupted	-2798
+▁depending	-2799
+▁effectual	-2800
+▁exception	-2801
+▁fortitude	-2802
+▁happeneth	-2803
+▁incessant	-2804
+▁ingenuous	-2805
+▁intellect	-2806
+▁neglected	-2807
+▁ourselves	-2808
+▁performed	-2809
+▁privilege	-2810
+▁qualities	-2811
+▁religious	-2812
+▁remaining	-2813
+▁requisite	-2814
+▁something	-2815
+▁sometimes	-2816
+▁throughly	-2817
+▁unlearned	-2818
+▁appetitive	-2819
+▁beneficial	-2820
+▁compounded	-2821
+▁compulsion	-2822
+▁conditions	-2823
+▁considered	-2824
+▁constantly	-2825
+▁determined	-2826
+▁distempers	-2827
+▁distracted	-2828
+▁endeavours	-2829
+▁excrements	-2830
+▁exhalation	-2831
+▁friendship	-2832
+▁generously	-2833
+▁hindrances	-2834
+▁moderation	-2835
+▁opposition	-2836
+▁physicians	-2837
+▁remembered	-2838
+▁ridiculous	-2839
+▁succession	-2840
+▁suspicious	-2841
+▁temperance	-2842
+▁thoroughly	-2843
+▁unsensible	-2844
+▁apprehended	-2845
+▁circumspect	-2846
+▁constrained	-2847
+▁contemplate	-2848
+▁contentedly	-2849
+▁distraction	-2850
+▁imagination	-2851
+▁opportunity	-2852
+▁particulars	-2853
+▁professions	-2854
+▁resolutions	-2855
+▁unnecessary	-2856
+▁wheresoever	-2857
+▁cheerfulness	-2858
+▁circumscribe	-2859
+▁commendation	-2860
+▁contemptible	-2861
+▁preservation	-2862
+▁concatenation	-2863
+▁ratiocination	-2864
+!)	-2865
+--	-2866
+;'	-2867
+AS	-2868
+Wh	-2869
+af	-2870
+ai	-2871
+aj	-2872
+ew	-2873
+gr	-2874
+ia	-2875
+lt	-2876
+yl	-2877
+The	-2878
+acy	-2879
+ali	-2880
+ape	-2881
+arb	-2882
+bus	-2883
+cho	-2884
+day	-2885
+eck	-2886
+ges	-2887
+get	-2888
+inc	-2889
+int	-2890
+isy	-2891
+its	-2892
+kon	-2893
+mpt	-2894
+nat	-2895
+nch	-2896
+nor	-2897
+nox	-2898
+ook	-2899
+oph	-2900
+ray	-2901
+ref	-2902
+rul	-2903
+see	-2904
+tak	-2905
+ued	-2906
+ung	-2907
+vet	-2908
+▁Al	-2909
+▁Am	-2910
+▁Cr	-2911
+▁Go	-2912
+▁On	-2913
+▁Sh	-2914
+▁We	-2915
+▁ca	-2916
+▁pa	-2917
+▁ra	-2918
+▁ye	-2919
+What	-2920
+aded	-2921
+ague	-2922
+airs	-2923
+aled	-2924
+aris	-2925
+aste	-2926
+atal	-2927
+atre	-2928
+atul	-2929
+aven	-2930
+cass	-2931
+ceit	-2932
+come	-2933
+edom	-2934
+ento	-2935
+erul	-2936
+fest	-2937
+geth	-2938
+gree	-2939
+hold	-2940
+iest	-2941
+ilty	-2942
+imon	-2943
+iots	-2944
+ised	-2945
+ites	-2946
+lict	-2947
+ners	-2948
+neth	-2949
+omes	-2950
+oney	-2951
+onst	-2952
+oubt	-2953
+ouch	-2954
+pass	-2955
+poll	-2956
+reme	-2957
+soci	-2958
+sper	-2959
+thou	-2960
+told	-2961
+ualt	-2962
+unct	-2963
+well	-2964
+welt	-2965
+▁Gre	-2966
+▁Per	-2967
+▁Sto	-2968
+▁Who	-2969
+▁aim	-2970
+▁ann	-2971
+▁arg	-2972
+▁asu	-2973
+▁bef	-2974
+▁bit	-2975
+▁bro	-2976
+▁cra	-2977
+▁dan	-2978
+▁der	-2979
+▁dev	-2980
+▁dom	-2981
+▁enm	-2982
+▁ens	-2983
+▁fil	-2984
+▁got	-2985
+▁hid	-2986
+▁irr	-2987
+▁jur	-2988
+▁lam	-2989
+▁nay	-2990
+▁pol	-2991
+▁reb	-2992
+▁rev	-2993
+▁rip	-2994
+▁saw	-2995
+▁sor	-2996
+▁sph	-2997
+▁top	-2998
+▁ver	-2999
+▁vil	-3000
+acles	-3001
+arest	-3002
+ators	-3003
+attle	-3004
+bable	-3005
+eling	-3006
+erald	-3007
+error	-3008
+eties	-3009
+iable	-3010
+iding	-3011
+ified	-3012
+igned	-3013
+illed	-3014
+inent	-3015
+irdly	-3016
+irous	-3017
+ister	-3018
+itute	-3019
+llect	-3020
+nexed	-3021
+odate	-3022
+oices	-3023
+olved	-3024
+oster	-3025
+pared	-3026
+quent	-3027
+rease	-3028
+right	-3029
+saith	-3030
+sseth	-3031
+ucing	-3032
+urren	-3033
+▁Chry	-3034
+▁Four	-3035
+▁Hath	-3036
+▁Hipp	-3037
+▁Look	-3038
+▁Pass	-3039
+▁Phil	-3040
+▁Some	-3041
+▁agre	-3042
+▁ball	-3043
+▁barb	-3044
+▁bere	-3045
+▁blow	-3046
+▁bold	-3047
+▁chie	-3048
+▁coat	-3049
+▁died	-3050
+▁dish	-3051
+▁duly	-3052
+▁dyed	-3053
+▁each	-3054
+▁envy	-3055
+▁face	-3056
+▁fact	-3057
+▁fail	-3058
+▁fine	-3059
+▁fish	-3060
+▁fits	-3061
+▁fond	-3062
+▁foot	-3063
+▁furn	-3064
+▁gift	-3065
+▁gold	-3066
+▁hair	-3067
+▁half	-3068
+▁herd	-3069
+▁holy	-3070
+▁home	-3071
+▁inqu	-3072
+▁king	-3073
+▁miss	-3074
+▁mixt	-3075
+▁mour	-3076
+▁neat	-3077
+▁omin	-3078
+▁pere	-3079
+▁plan	-3080
+▁poor	-3081
+▁poss	-3082
+▁pret	-3083
+▁quen	-3084
+▁rank	-3085
+▁reun	-3086
+▁rich	-3087
+▁skin	-3088
+▁slow	-3089
+▁sold	-3090
+▁stor	-3091
+▁talk	-3092
+▁toil	-3093
+▁vari	-3094
+▁wait	-3095
+▁went	-3096
+▁womb	-3097
+▁wood	-3098
+asting	-3099
+cience	-3100
+ensive	-3101
+estics	-3102
+etoric	-3103
+iction	-3104
+itness	-3105
+jected	-3106
+morrow	-3107
+mptory	-3108
+phists	-3109
+ressed	-3110
+ruling	-3111
+▁Athen	-3112
+▁Death	-3113
+▁Every	-3114
+▁Gover	-3115
+▁Never	-3116
+▁Whose	-3117
+▁assid	-3118
+▁baths	-3119
+▁begot	-3120
+▁below	-3121
+▁blind	-3122
+▁books	-3123
+▁brave	-3124
+▁chaos	-3125
+▁check	-3126
+▁cloth	-3127
+▁covet	-3128
+▁cross	-3129
+▁curse	-3130
+▁deser	-3131
+▁disci	-3132
+▁doubt	-3133
+▁drink	-3134
+▁dwelt	-3135
+▁dying	-3136
+▁eaten	-3137
+▁fatal	-3138
+▁field	-3139
+▁fight	-3140
+▁fixed	-3141
+▁gives	-3142
+▁grand	-3143
+▁greed	-3144
+▁guide	-3145
+▁habit	-3146
+▁hours	-3147
+▁joint	-3148
+▁liver	-3149
+▁meats	-3150
+▁meets	-3151
+▁mercy	-3152
+▁obnox	-3153
+▁poets	-3154
+▁prove	-3155
+▁pursu	-3156
+▁seeds	-3157
+▁seeks	-3158
+▁share	-3159
+▁smell	-3160
+▁spare	-3161
+▁spect	-3162
+▁stars	-3163
+▁steal	-3164
+▁stone	-3165
+▁swift	-3166
+▁trade	-3167
+▁tresp	-3168
+▁uncon	-3169
+▁union	-3170
+▁useth	-3171
+▁utter	-3172
+▁vexed	-3173
+▁voice	-3174
+▁woman	-3175
+▁wound	-3176
+animous	-3177
+ariwise	-3178
+attered	-3179
+ensible	-3180
+ishment	-3181
+justice	-3182
+ocrates	-3183
+owardly	-3184
+ression	-3185
+usticus	-3186
+willing	-3187
+▁Behold	-3188
+▁Within	-3189
+▁accomm	-3190
+▁adhere	-3191
+▁admire	-3192
+▁attend	-3193
+▁bitten	-3194
+▁branch	-3195
+▁carnal	-3196
+▁choice	-3197
+▁cities	-3198
+▁cleave	-3199
+▁comedy	-3200
+▁degree	-3201
+▁determ	-3202
+▁dispra	-3203
+▁family	-3204
+▁fitted	-3205
+▁forced	-3206
+▁foster	-3207
+▁fruits	-3208
+▁gather	-3209
+▁goodly	-3210
+▁hasten	-3211
+▁heaven	-3212
+▁idiots	-3213
+▁jurisd	-3214
+▁lawful	-3215
+▁nearer	-3216
+▁needed	-3217
+▁nobody	-3218
+▁obtain	-3219
+▁oppose	-3220
+▁parent	-3221
+▁partic	-3222
+▁pierce	-3223
+▁plague	-3224
+▁poetry	-3225
+▁prince	-3226
+▁purple	-3227
+▁putref	-3228
+▁putrid	-3229
+▁querul	-3230
+▁really	-3231
+▁reluct	-3232
+▁remiss	-3233
+▁sacred	-3234
+▁sacrif	-3235
+▁sayest	-3236
+▁search	-3237
+▁seemed	-3238
+▁settle	-3239
+▁sinned	-3240
+▁speaks	-3241
+▁sphere	-3242
+▁taught	-3243
+▁tenets	-3244
+▁termed	-3245
+▁termin	-3246
+▁terror	-3247
+▁thrown	-3248
+▁tracts	-3249
+▁transl	-3250
+▁turned	-3251
+▁tyrant	-3252
+amations	-3253
+ertainly	-3254
+fferency	-3255
+ittingly	-3256
+ordinary	-3257
+ountains	-3258
+robation	-3259
+urrences	-3260
+▁Nothing	-3261
+▁Thirdly	-3262
+▁Wherein	-3263
+▁admired	-3264
+▁affairs	-3265
+▁allowed	-3266
+▁annexed	-3267
+▁applied	-3268
+▁approve	-3269
+▁asunder	-3270
+▁becomes	-3271
+▁calling	-3272
+▁carcass	-3273
+▁casualt	-3274
+▁challen	-3275
+▁channel	-3276
+▁commits	-3277
+▁company	-3278
+▁condemn	-3279
+▁confess	-3280
+▁confine	-3281
+▁conform	-3282
+▁contest	-3283
+▁couldst	-3284
+▁customs	-3285
+▁dealing	-3286
+▁decline	-3287
+▁desired	-3288
+▁destiny	-3289
+▁discour	-3290
+▁dispens	-3291
+▁element	-3292
+▁emerald	-3293
+▁essence	-3294
+▁exclaim	-3295
+▁exclude	-3296
+▁falling	-3297
+▁feeling	-3298
+▁foresee	-3299
+▁freedom	-3300
+▁grounds	-3301
+▁happily	-3302
+▁hearken	-3303
+▁imitate	-3304
+▁letters	-3305
+▁memento	-3306
+▁mixture	-3307
+▁offends	-3308
+▁ominous	-3309
+▁operate	-3310
+▁opposed	-3311
+▁periods	-3312
+▁prevail	-3313
+▁project	-3314
+▁prosper	-3315
+▁reading	-3316
+▁redress	-3317
+▁reflect	-3318
+▁seemeth	-3319
+▁serious	-3320
+▁servile	-3321
+▁sinneth	-3322
+▁slander	-3323
+▁station	-3324
+▁stories	-3325
+▁therein	-3326
+▁thereof	-3327
+▁tongues	-3328
+▁turneth	-3329
+▁usually	-3330
+▁willing	-3331
+▁worship	-3332
+▁Governor	-3333
+▁admitted	-3334
+▁calamity	-3335
+▁chiefest	-3336
+▁conceits	-3337
+▁confined	-3338
+▁confused	-3339
+▁consumed	-3340
+▁cowardly	-3341
+▁dejected	-3342
+▁desirous	-3343
+▁difficul	-3344
+▁equality	-3345
+▁examples	-3346
+▁foretold	-3347
+▁governor	-3348
+▁involved	-3349
+▁kindness	-3350
+▁lovingly	-3351
+▁meditate	-3352
+▁miseries	-3353
+▁mistress	-3354
+▁moderate	-3355
+▁moisture	-3356
+▁obedient	-3357
+▁overcome	-3358
+▁partakes	-3359
+▁perished	-3360
+▁persuade	-3361
+▁pleaseth	-3362
+▁possibly	-3363
+▁prepared	-3364
+▁probable	-3365
+▁proposed	-3366
+▁prudence	-3367
+▁rashness	-3368
+▁recourse	-3369
+▁retiring	-3370
+▁reunited	-3371
+▁rhetoric	-3372
+▁secretly	-3373
+▁stranger	-3374
+▁stronger	-3375
+▁surprise	-3376
+▁thinking	-3377
+▁transfer	-3378
+▁trespass	-3379
+▁Certainly	-3380
+▁accounted	-3381
+▁agreeably	-3382
+▁appetites	-3383
+▁authority	-3384
+▁belonging	-3385
+▁bountiful	-3386
+▁cessation	-3387
+▁conceited	-3388
+▁conducing	-3389
+▁confident	-3390
+▁constancy	-3391
+▁contained	-3392
+▁continued	-3393
+▁diligence	-3394
+▁dishonour	-3395
+▁dissemble	-3396
+▁distribut	-3397
+▁domestics	-3398
+▁efficient	-3399
+▁excessive	-3400
+▁existence	-3401
+▁greatness	-3402
+▁hypocrisy	-3403
+▁imperfect	-3404
+▁injustice	-3405
+▁intercept	-3406
+▁malicious	-3407
+▁mischance	-3408
+▁mountains	-3409
+▁multitude	-3410
+▁murmuring	-3411
+▁mysteries	-3412
+▁obnoxious	-3413
+▁offensive	-3414
+▁perfectly	-3415
+▁prejudice	-3416
+▁privately	-3417
+▁prosecute	-3418
+▁recollect	-3419
+▁remembers	-3420
+▁respected	-3421
+▁scattered	-3422
+▁suspected	-3423
+▁temperate	-3424
+▁thousands	-3425
+▁uncertain	-3426
+▁unwilling	-3427
+▁acceptable	-3428
+▁acquainted	-3429
+▁attainment	-3430
+▁commanding	-3431
+▁conscience	-3432
+▁deliberate	-3433
+▁delightful	-3434
+▁equanimity	-3435
+▁extinction	-3436
+▁familiarly	-3437
+▁grievously	-3438
+▁incidental	-3439
+▁insensible	-3440
+▁irrational	-3441
+▁maintained	-3442
+▁peremptory	-3443
+▁proceeding	-3444
+▁properties	-3445
+▁propitious	-3446
+▁proportion	-3447
+▁prosecuted	-3448
+▁reputation	-3449
+▁separation	-3450
+▁succeedeth	-3451
+▁accommodate	-3452
+▁accordingly	-3453
+▁complaining	-3454
+▁composition	-3455
+▁examination	-3456
+▁impediments	-3457
+▁interrupted	-3458
+▁occurrences	-3459
+▁remembering	-3460
+▁respiration	-3461
+▁subordinate	-3462
+▁comprehended	-3463
+▁consequences	-3464
+▁continuation	-3465
+▁contrariwise	-3466
+▁indifferency	-3467
+▁instructions	-3468
+▁jurisdiction	-3469
+▁praiseworthy	-3470
+▁putrefaction	-3471
+▁uncharitable	-3472
+▁extraordinary	-3473
+▁indifferently	-3474
+▁understandeth	-3475
+▁administration	-3476
+▁circumspection	-3477
+):	-3478
+HE	-3479
+In	-3480
+eg	-3481
+fl	-3482
+go	-3483
+ih	-3484
+...	-3485
+AST	-3486
+And	-3487
+But	-3488
+God	-3489
+III	-3490
+acr	-3491
+adf	-3492
+als	-3493
+arn	-3494
+aus	-3495
+bri	-3496
+cul	-3497
+els	-3498
+eme	-3499
+.,	-3500
+kn	-3501
+arm	-3502
+equ	-3503
+ext	-3504
+fid	-3505
+fty	-3506
+ids	-3507
+ike	-3508
+iog	-3509
+isp	-3510
+jur	-3511
+lag	-3512
+lem	-3513
+log	-3514
+lot	-3515
+lux	-3516
+man	-3517
+mos	-3518
+ocr	-3519
+ops	-3520
+orn	-3521
+ott	-3522
+ped	-3523
+pro	-3524
+ram	-3525
+rus	-3526
+sar	-3527
+sim	-3528
+sin	-3529
+sue	-3530
+ugg	-3531
+ulf	-3532
+val	-3533
+vel	-3534
+vey	-3535
+yth	-3536
+▁Ar	-3537
+▁Ca	-3538
+▁Em	-3539
+▁Mu	-3540
+▁My	-3541
+▁Oh	-3542
+▁St	-3543
+They	-3544
+abul	-3545
+aced	-3546
+ages	-3547
+agog	-3548
+aine	-3549
+ains	-3550
+alog	-3551
+ancr	-3552
+anic	-3553
+ansh	-3554
+anth	-3555
+arel	-3556
+asis	-3557
+aton	-3558
+aund	-3559
+aver	-3560
+bber	-3561
+cers	-3562
+cure	-3563
+eace	-3564
+eper	-3565
+erus	-3566
+esty	-3567
+etri	-3568
+etty	-3569
+fish	-3570
+hood	-3571
+iery	-3572
+ihil	-3573
+illa	-3574
+ilst	-3575
+inke	-3576
+inue	-3577
+ires	-3578
+ises	-3579
+ison	-3580
+istr	-3581
+iter	-3582
+ithe	-3583
+itic	-3584
+ivel	-3585
+know	-3586
+mast	-3587
+odot	-3588
+oles	-3589
+oman	-3590
+omed	-3591
+omin	-3592
+ompl	-3593
+onum	-3594
+ored	-3595
+outh	-3596
+oved	-3597
+pent	-3598
+pper	-3599
+;)	-3600
+My	-3601
+ob	-3602
+irm	-3603
+nda	-3604
+nic	-3605
+roy	-3606
+urb	-3607
+▁ak	-3608
+▁av	-3609
+▁ax	-3610
+▁ly	-3611
+▁ox	-3612
+▁sn	-3613
+aind	-3614
+arer	-3615
+crat	-3616
+enty	-3617
+ingu	-3618
+ling	-3619
+rary	-3620
+rase	-3621
+rief	-3622
+ries	-3623
+rily	-3624
+ript	-3625
+roan	-3626
+roat	-3627
+said	-3628
+sels	-3629
+side	-3630
+stro	-3631
+that	-3632
+ting	-3633
+uded	-3634
+udit	-3635
+ules	-3636
+uman	-3637
+umes	-3638
+unda	-3639
+upid	-3640
+ural	-3641
+ures	-3642
+vell	-3643
+when	-3644
+▁Are	-3645
+▁Dio	-3646
+▁Eud	-3647
+▁Fab	-3648
+▁Fif	-3649
+▁Hel	-3650
+▁Luc	-3651
+▁Med	-3652
+▁Nay	-3653
+▁Our	-3654
+▁She	-3655
+▁THE	-3656
+▁Tra	-3657
+▁Was	-3658
+▁Wor	-3659
+▁ant	-3660
+▁anx	-3661
+▁apt	-3662
+▁bew	-3663
+▁civ	-3664
+▁clo	-3665
+▁cou	-3666
+▁cry	-3667
+▁cyl	-3668
+▁den	-3669
+▁elo	-3670
+▁emb	-3671
+▁fet	-3672
+▁hap	-3673
+▁hot	-3674
+▁hou	-3675
+▁kos	-3676
+▁lar	-3677
+▁lod	-3678
+▁log	-3679
+▁maj	-3680
+▁met	-3681
+▁oli	-3682
+▁par	-3683
+▁ped	-3684
+▁pil	-3685
+▁pit	-3686
+▁pri	-3687
+▁rig	-3688
+▁sed	-3689
+▁sei	-3690
+▁shr	-3691
+▁sil	-3692
+▁sla	-3693
+▁sto	-3694
+▁sup	-3695
+▁unf	-3696
+▁ves	-3697
+▁wax	-3698
+▁wil	-3699
+my	-3700
+git	-3701
+pon	-3702
+ters	-3703
+▁lay	-3704
+▁owe	-3705
+▁pam	-3706
+abled	-3707
+acies	-3708
+adful	-3709
+ailty	-3710
+andle	-3711
+angle	-3712
+asing	-3713
+asion	-3714
+athos	-3715
+aving	-3716
+baine	-3717
+bling	-3718
+cepts	-3719
+chool	-3720
+eason	-3721
+enced	-3722
+equal	-3723
+erest	-3724
+essed	-3725
+gancy	-3726
+gging	-3727
+ially	-3728
+icacy	-3729
+ickle	-3730
+iders	-3731
+imony	-3732
+inius	-3733
+ionat	-3734
+isdom	-3735
+ising	-3736
+ision	-3737
+iving	-3738
+losed	-3739
+mansh	-3740
+ocles	-3741
+omach	-3742
+ombus	-3743
+onius	-3744
+oping	-3745
+order	-3746
+ordes	-3747
+orous	-3748
+osses	-3749
+pered	-3750
+poral	-3751
+rible	-3752
+rious	-3753
+roper	-3754
+sease	-3755
+stain	-3756
+ttled	-3757
+ulent	-3758
+utish	-3759
+vered	-3760
+whose	-3761
+ythag	-3762
+▁Away	-3763
+▁Cato	-3764
+▁Char	-3765
+▁Diog	-3766
+▁Diot	-3767
+▁Euty	-3768
+▁Even	-3769
+▁Give	-3770
+▁Have	-3771
+▁LAST	-3772
+▁Lord	-3773
+▁Many	-3774
+▁Next	-3775
+▁Only	-3776
+▁Phal	-3777
+▁Seen	-3778
+▁Wipe	-3779
+▁accl	-3780
+▁afar	-3781
+▁amph	-3782
+▁ants	-3783
+▁aust	-3784
+▁beds	-3785
+▁bees	-3786
+▁boar	-3787
+▁brow	-3788
+▁capt	-3789
+▁clod	-3790
+▁cold	-3791
+▁conv	-3792
+▁corn	-3793
+▁cott	-3794
+▁date	-3795
+▁diet	-3796
+▁dign	-3797
+▁dirt	-3798
+▁dism	-3799
+▁dram	-3800
+▁draw	-3801
+▁dung	-3802
+▁exhi	-3803
+▁exor	-3804
+▁fate	-3805
+▁feed	-3806
+▁fest	-3807
+▁fulf	-3808
+▁heap	-3809
+▁hero	-3810
+▁hill	-3811
+▁imit	-3812
+▁impr	-3813
+▁insp	-3814
+▁irre	-3815
+▁kill	-3816
+���knew	-3817
+▁knit	-3818
+▁land	-3819
+▁laws	-3820
+▁leaf	-3821
+▁lets	-3822
+▁lies	-3823
+▁line	-3824
+▁lion	-3825
+▁lost	-3826
+▁mech	-3827
+▁nour	-3828
+▁obso	-3829
+▁pity	-3830
+▁post	-3831
+▁prom	-3832
+▁pron	-3833
+▁pupp	-3834
+▁rage	-3835
+▁remn	-3836
+▁robe	-3837
+▁roll	-3838
+▁rove	-3839
+▁rude	-3840
+▁runs	-3841
+▁sees	-3842
+▁sent	-3843
+▁song	-3844
+▁sore	-3845
+▁stre	-3846
+▁summ	-3847
+▁tame	-3848
+▁test	-3849
+▁trad	-3850
+▁tune	-3851
+▁warm	-3852
+▁wars	-3853
+▁weak	-3854
+▁wipe	-3855
+▁wish	-3856
+▁your	-3857
+abited	-3858
+acrity	-3859
+agated	-3860
+ainder	-3861
+arians	-3862
+athing	-3863
+atiast	-3864
+atonic	-3865
+attern	-3866
+briety	-3867
+cerity	-3868
+ecause	-3869
+either	-3870
+escend	-3871
+esides	-3872
+etrius	-3873
+everus	-3874
+fected	-3875
+iguity	-3876
+inally	-3877
+indrus	-3878
+inency	-3879
+inkest	-3880
+isibly	-3881
+isseth	-3882
+itteth	-3883
+itynch	-3884
+lessed	-3885
+lestra	-3886
+liness	-3887
+odotus	-3888
+ompreh	-3889
+ongeth	-3890
+ounter	-3891
+oveful	-3892
+penter	-3893
+remity	-3894
+rified	-3895
+rition	-3896
+sences	-3897
+sippus	-3898
+tances	-3899
+tofore	-3900
+uously	-3901
+usions	-3902
+verted	-3903
+▁Admin	-3904
+▁Among	-3905
+▁Apoll	-3906
+▁Caius	-3907
+▁Catul	-3908
+▁Emped	-3909
+▁Fancy	-3910
+▁Gener	-3911
+▁Greci	-3912
+▁Muses	-3913
+▁Seest	-3914
+▁Shall	-3915
+▁Stoic	-3916
+▁Their	-3917
+▁Think	-3918
+▁abuse	-3919
+▁accur	-3920
+▁accus	-3921
+▁acute	-3922
+▁alive	-3923
+▁allay	-3924
+▁alleg	-3925
+▁along	-3926
+▁aside	-3927
+▁audit	-3928
+▁aught	-3929
+▁avail	-3930
+▁belly	-3931
+▁black	-3932
+▁bless	-3933
+▁board	-3934
+▁bones	-3935
+▁break	-3936
+▁brief	-3937
+▁brook	-3938
+▁brute	-3939
+▁build	-3940
+▁	-3941
+e	-3942
+t	-3943
+o	-3944
+a	-3945
+n	-3946
+h	-3947
+i	-3948
+s	-3949
+r	-3950
+d	-3951
+l	-3952
+u	-3953
+f	-3954
+c	-3955
+m	-3956
+,	-3957
+y	-3958
+w	-3959
+p	-3960
+b	-3961
+g	-3962
+v	-3963
+.	-3964
+k	-3965
+;	-3966
+I	-3967
+:	-3968
+?	-3969
+A	-3970
+T	-3971
+x	-3972
+j	-3973
+F	-3974
+W	-3975
+'	-3976
+q	-3977
+B	-3978
+(	-3979
+)	-3980
+H	-3981
+S	-3982
+G	-3983
+O	-3984
+N	-3985
+-	-3986
+C	-3987
+L	-3988
+D	-3989
+P	-3990
+E	-3991
+M	-3992
+!	-3993

requirements.txt

@@ -0,0 +1,10 @@
+numpy
+torch
+sentencepiece
+openai
+python-dotenv
+unsloth
+datasets
+trl
+transformers
+argparse