Implement initial project structure and setup

runs/neural_chonky_distilbert_base_uncased_1_A_Christmas_Carol_in_Prose_Bei.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.13793103448275862,
+      "3": 0.2413793103448276,
+      "5": 0.2413793103448276,
+      "10": 0.27586206896551724
+    }
+  },
+  "metadata": {
+    "chunker_type": "NeuralChunker",
+    "chunker_config": "NeuralChunker(model=DistilBertForTokenClassification(\n  (distilbert): DistilBertModel(\n    (embeddings): Embeddings(\n      (word_embeddings): Embedding(30522, 768, padding_idx=0)\n      (position_embeddings): Embedding(512, 768)\n      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n      (dropout): Dropout(p=0.1, inplace=False)\n    )\n    (transformer): Transformer(\n      (layer): ModuleList(\n        (0-5): 6 x TransformerBlock(\n          (attention): DistilBertSdpaAttention(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (q_lin): Linear(in_features=768, out_features=768, bias=True)\n            (k_lin): Linear(in_features=768, out_features=768, bias=True)\n            (v_lin): Linear(in_features=768, out_features=768, bias=True)\n            (out_lin): Linear(in_features=768, out_features=768, bias=True)\n          )\n          (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n          (ffn): FFN(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (lin1): Linear(in_features=768, out_features=3072, bias=True)\n            (lin2): Linear(in_features=3072, out_features=768, bias=True)\n            (activation): GELUActivation()\n          )\n          (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n        )\n      )\n    )\n  )\n  (dropout): Dropout(p=0.1, inplace=False)\n  (classifier): Linear(in_features=768, out_features=2, bias=True)\n),tokenizer=<chonkie.tokenizer.Tokenizer object at 0x11a8b0200>, min_characters_per_chunk=10, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x34cb67fe0>",
+    "total_questions": 29,
+    "total_correct": {
+      "1": 4,
+      "3": 7,
+      "5": 7,
+      "10": 8
+    }
+  },
+  "total_corpus_size_mb": 0.15099716186523438,
+  "total_time_to_chunk": 5.525778293609619,
+  "chunk_speed_mb_per_sec": 0.02732595371042259,
+  "total_chunks_created": 416,
+  "total_evaluation_time": 5.571743011474609,
+  "questions_per_second": 5.2048344549051455,
+  "timestamp": 1750445372.16346
+}

runs/neuralchunker_A_Christmas_Carol_in_Prose_Bei.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.13793103448275862,
+      "3": 0.2413793103448276,
+      "5": 0.2413793103448276,
+      "10": 0.27586206896551724
+    }
+  },
+  "metadata": {
+    "chunker_type": "NeuralChunker",
+    "chunker_config": "NeuralChunker(model=DistilBertForTokenClassification(\n  (distilbert): DistilBertModel(\n    (embeddings): Embeddings(\n      (word_embeddings): Embedding(30522, 768, padding_idx=0)\n      (position_embeddings): Embedding(512, 768)\n      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n      (dropout): Dropout(p=0.1, inplace=False)\n    )\n    (transformer): Transformer(\n      (layer): ModuleList(\n        (0-5): 6 x TransformerBlock(\n          (attention): DistilBertSdpaAttention(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (q_lin): Linear(in_features=768, out_features=768, bias=True)\n            (k_lin): Linear(in_features=768, out_features=768, bias=True)\n            (v_lin): Linear(in_features=768, out_features=768, bias=True)\n            (out_lin): Linear(in_features=768, out_features=768, bias=True)\n          )\n          (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n          (ffn): FFN(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (lin1): Linear(in_features=768, out_features=3072, bias=True)\n            (lin2): Linear(in_features=3072, out_features=768, bias=True)\n            (activation): GELUActivation()\n          )\n          (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n        )\n      )\n    )\n  )\n  (dropout): Dropout(p=0.1, inplace=False)\n  (classifier): Linear(in_features=768, out_features=2, bias=True)\n),tokenizer=<chonkie.tokenizer.Tokenizer object at 0x112060620>, min_characters_per_chunk=10, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x168f57fe0>",
+    "total_questions": 29,
+    "total_correct": {
+      "1": 4,
+      "3": 7,
+      "5": 7,
+      "10": 8
+    }
+  },
+  "total_corpus_size_mb": 0.15099716186523438,
+  "total_time_to_chunk": 5.5413689613342285,
+  "chunk_speed_mb_per_sec": 0.02724907201069641,
+  "total_chunks_created": 416,
+  "total_evaluation_time": 5.585696697235107,
+  "questions_per_second": 5.191832204987939,
+  "timestamp": 1750444854.788678
+}

runs/neuralchunker_Daddy_Takes_Us_to_the_Garden.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.20689655172413793,
+      "3": 0.4827586206896552,
+      "5": 0.4827586206896552,
+      "10": 0.5172413793103449
+    }
+  },
+  "metadata": {
+    "chunker_type": "NeuralChunker",
+    "chunker_config": "NeuralChunker(model=DistilBertForTokenClassification(\n  (distilbert): DistilBertModel(\n    (embeddings): Embeddings(\n      (word_embeddings): Embedding(30522, 768, padding_idx=0)\n      (position_embeddings): Embedding(512, 768)\n      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n      (dropout): Dropout(p=0.1, inplace=False)\n    )\n    (transformer): Transformer(\n      (layer): ModuleList(\n        (0-5): 6 x TransformerBlock(\n          (attention): DistilBertSdpaAttention(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (q_lin): Linear(in_features=768, out_features=768, bias=True)\n            (k_lin): Linear(in_features=768, out_features=768, bias=True)\n            (v_lin): Linear(in_features=768, out_features=768, bias=True)\n            (out_lin): Linear(in_features=768, out_features=768, bias=True)\n          )\n          (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n          (ffn): FFN(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (lin1): Linear(in_features=768, out_features=3072, bias=True)\n            (lin2): Linear(in_features=3072, out_features=768, bias=True)\n            (activation): GELUActivation()\n          )\n          (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n        )\n      )\n    )\n  )\n  (dropout): Dropout(p=0.1, inplace=False)\n  (classifier): Linear(in_features=768, out_features=2, bias=True)\n),tokenizer=<chonkie.tokenizer.Tokenizer object at 0x119462e10>, min_characters_per_chunk=10, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x16d43ffe0>",
+    "total_questions": 29,
+    "total_correct": {
+      "1": 6,
+      "3": 14,
+      "5": 14,
+      "10": 15
+    }
+  },
+  "total_corpus_size_mb": 0.15877437591552734,
+  "total_time_to_chunk": 5.989755868911743,
+  "chunk_speed_mb_per_sec": 0.026507653966266656,
+  "total_chunks_created": 502,
+  "total_evaluation_time": 6.037913084030151,
+  "questions_per_second": 4.802984010601763,
+  "timestamp": 1750444663.594319
+}

runs/neuralchunker_Riders_of_the_Silences.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.2857142857142857,
+      "3": 0.39285714285714285,
+      "5": 0.4642857142857143,
+      "10": 0.5714285714285714
+    }
+  },
+  "metadata": {
+    "chunker_type": "NeuralChunker",
+    "chunker_config": "NeuralChunker(model=DistilBertForTokenClassification(\n  (distilbert): DistilBertModel(\n    (embeddings): Embeddings(\n      (word_embeddings): Embedding(30522, 768, padding_idx=0)\n      (position_embeddings): Embedding(512, 768)\n      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n      (dropout): Dropout(p=0.1, inplace=False)\n    )\n    (transformer): Transformer(\n      (layer): ModuleList(\n        (0-5): 6 x TransformerBlock(\n          (attention): DistilBertSdpaAttention(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (q_lin): Linear(in_features=768, out_features=768, bias=True)\n            (k_lin): Linear(in_features=768, out_features=768, bias=True)\n            (v_lin): Linear(in_features=768, out_features=768, bias=True)\n            (out_lin): Linear(in_features=768, out_features=768, bias=True)\n          )\n          (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n          (ffn): FFN(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (lin1): Linear(in_features=768, out_features=3072, bias=True)\n            (lin2): Linear(in_features=3072, out_features=768, bias=True)\n            (activation): GELUActivation()\n          )\n          (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n        )\n      )\n    )\n  )\n  (dropout): Dropout(p=0.1, inplace=False)\n  (classifier): Linear(in_features=768, out_features=2, bias=True)\n),tokenizer=<chonkie.tokenizer.Tokenizer object at 0x1625750a0>, min_characters_per_chunk=10, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x1646d3fe0>",
+    "total_questions": 28,
+    "total_correct": {
+      "1": 8,
+      "3": 11,
+      "5": 13,
+      "10": 16
+    }
+  },
+  "total_corpus_size_mb": 0.3561391830444336,
+  "total_time_to_chunk": 12.656232357025146,
+  "chunk_speed_mb_per_sec": 0.028139431467276276,
+  "total_chunks_created": 858,
+  "total_evaluation_time": 12.763389825820923,
+  "questions_per_second": 2.193774567893767,
+  "timestamp": 1750444825.624423
+}

runs/neuralchunker_The_Reign_of_Greed.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.3333333333333333,
+      "3": 0.4666666666666667,
+      "5": 0.5,
+      "10": 0.6333333333333333
+    }
+  },
+  "metadata": {
+    "chunker_type": "NeuralChunker",
+    "chunker_config": "NeuralChunker(model=DistilBertForTokenClassification(\n  (distilbert): DistilBertModel(\n    (embeddings): Embeddings(\n      (word_embeddings): Embedding(30522, 768, padding_idx=0)\n      (position_embeddings): Embedding(512, 768)\n      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n      (dropout): Dropout(p=0.1, inplace=False)\n    )\n    (transformer): Transformer(\n      (layer): ModuleList(\n        (0-5): 6 x TransformerBlock(\n          (attention): DistilBertSdpaAttention(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (q_lin): Linear(in_features=768, out_features=768, bias=True)\n            (k_lin): Linear(in_features=768, out_features=768, bias=True)\n            (v_lin): Linear(in_features=768, out_features=768, bias=True)\n            (out_lin): Linear(in_features=768, out_features=768, bias=True)\n          )\n          (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n          (ffn): FFN(\n            (dropout): Dropout(p=0.1, inplace=False)\n            (lin1): Linear(in_features=768, out_features=3072, bias=True)\n            (lin2): Linear(in_features=3072, out_features=768, bias=True)\n            (activation): GELUActivation()\n          )\n          (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n        )\n      )\n    )\n  )\n  (dropout): Dropout(p=0.1, inplace=False)\n  (classifier): Linear(in_features=768, out_features=2, bias=True)\n),tokenizer=<chonkie.tokenizer.Tokenizer object at 0x142bf9100>, min_characters_per_chunk=10, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x145457fe0>",
+    "total_questions": 30,
+    "total_correct": {
+      "1": 10,
+      "3": 14,
+      "5": 15,
+      "10": 19
+    }
+  },
+  "total_corpus_size_mb": 0.6241388320922852,
+  "total_time_to_chunk": 20.625742197036743,
+  "chunk_speed_mb_per_sec": 0.030260187785240224,
+  "total_chunks_created": 1284,
+  "total_evaluation_time": 20.78759002685547,
+  "questions_per_second": 1.4431687348674391,
+  "timestamp": 1750444629.6744802
+}

runs/recursivechunker_1024.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.28120665742024964,
+      "3": 0.4604715672676838,
+      "5": 0.5319001386962552,
+      "10": 0.6324549237170597
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x106ebec00>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=1024, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11cb94b90>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 811,
+      "3": 1328,
+      "5": 1534,
+      "10": 1824
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.15627236366272,
+  "chunk_speed_mb_per_sec": 30.586160302442867,
+  "total_chunks_created": 18732,
+  "total_evaluation_time": 28.514283657073975,
+  "questions_per_second": 101.1422918662213,
+  "timestamp": 1750411622.8747
+}

runs/recursivechunker_128.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.21185852981969486,
+      "3": 0.32420249653259364,
+      "5": 0.37552011095700416,
+      "10": 0.45180305131761445
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x10bb350a0>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=128, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11c3dc0b0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 611,
+      "3": 935,
+      "5": 1083,
+      "10": 1303
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.8560697317123416,
+  "chunk_speed_mb_per_sec": 23.09190543858819,
+  "total_chunks_created": 190936,
+  "total_evaluation_time": 43.952011823654175,
+  "questions_per_second": 65.6170191155592,
+  "timestamp": 1750411518.560171
+}

runs/recursivechunker_2048.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.23370319001386963,
+      "3": 0.3699722607489598,
+      "5": 0.4573509015256588,
+      "10": 0.584257975034674
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x10be36510>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=2048, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11d35d220>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 674,
+      "3": 1067,
+      "5": 1319,
+      "10": 1685
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 1.9020262718200682,
+  "chunk_speed_mb_per_sec": 34.67464837255119,
+  "total_chunks_created": 8940,
+  "total_evaluation_time": 25.84335994720459,
+  "questions_per_second": 111.59539649224114,
+  "timestamp": 1750411651.5344021
+}

runs/recursivechunker_256.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.28398058252427183,
+      "3": 0.4240638002773925,
+      "5": 0.4934119278779473,
+      "10": 0.5856449375866851
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x1064caea0>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=256, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10bf4d8b0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 819,
+      "3": 1223,
+      "5": 1423,
+      "10": 1689
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.5457475900650026,
+  "chunk_speed_mb_per_sec": 25.90676798756444,
+  "total_chunks_created": 89001,
+  "total_evaluation_time": 35.80127787590027,
+  "questions_per_second": 80.55578379064991,
+  "timestamp": 1750411557.8257332
+}

runs/recursivechunker_3072.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.21151178918169208,
+      "3": 0.358876560332871,
+      "5": 0.45041608876560335,
+      "10": 0.5970873786407767
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x11ccd89e0>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=3072, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11cbcbef0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 610,
+      "3": 1035,
+      "5": 1299,
+      "10": 1722
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 1.8356848478317256,
+  "chunk_speed_mb_per_sec": 35.92778588798433,
+  "total_chunks_created": 5870,
+  "total_evaluation_time": 25.02839708328247,
+  "questions_per_second": 115.22911317106863,
+  "timestamp": 1750411679.182185
+}

runs/recursivechunker_512.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.32524271844660196,
+      "3": 0.4920249653259362,
+      "5": 0.5696948682385575,
+      "10": 0.6726768377253814
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x10bce61b0>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=512, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10bb11d60>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 938,
+      "3": 1419,
+      "5": 1643,
+      "10": 1940
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.342450857162475,
+  "chunk_speed_mb_per_sec": 28.155165761132046,
+  "total_chunks_created": 40552,
+  "total_evaluation_time": 30.905266761779785,
+  "questions_per_second": 93.31742781028547,
+  "timestamp": 1750411591.337388
+}

runs/recursivechunker_512_openai.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.47156726768377255,
+      "3": 0.6685159500693482,
+      "5": 0.7479195561719834,
+      "10": 0.8359916782246879
+    }
+  },
+  "metadata": {
+    "chunker_type": "RecursiveChunker",
+    "chunker_config": "RecursiveChunker(tokenizer_or_token_counter=<chonkie.tokenizer.Tokenizer object at 0x10435ed20>, rules=RecursiveRules(levels=[RecursiveLevel(delimiters=['\\n\\n', '\\r\\n', '\\n', '\\r'], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['. ', '! ', '? '], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=['{', '}', '\"', '[', ']', '<', '>', '(', ')', ':', ';', ',', '\u2014', '|', '~', '-', '...', '`', \"'\"], whitespace=False, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=True, include_delim=prev), RecursiveLevel(delimiters=None, whitespace=False, include_delim=prev)]), chunk_size=512, min_characters_per_chunk=24, return_type=chunks)",
+    "embedding_model": "OpenAIEmbeddings(model=text-embedding-3-small)",
+    "embedding_model_name": "text-embedding-3-small",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 1360,
+      "3": 1928,
+      "5": 2157,
+      "10": 2411
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 95.89966173171999,
+  "chunk_speed_mb_per_sec": 0.687719758128207,
+  "total_chunks_created": 40552,
+  "total_evaluation_time": 967.4273478984833,
+  "questions_per_second": 2.981102411736485,
+  "timestamp": 1750413470.612622
+}

runs/semanticchunker_default.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.13626907073509015,
+      "3": 0.22711511789181693,
+      "5": 0.2708044382801664,
+      "10": 0.3273231622746186
+    }
+  },
+  "metadata": {
+    "chunker_type": "SemanticChunker",
+    "chunker_config": "SemanticChunker(model=Model2VecEmbeddings(model=minishlab/potion-base-8M), chunk_size=512, mode=window, threshold=auto, similarity_window=1, min_sentences=1, min_chunk_size=2, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-base-8M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10e3c5040>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 393,
+      "3": 655,
+      "5": 781,
+      "10": 944
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 7.4704547166824335,
+  "chunk_speed_mb_per_sec": 8.828390596282754,
+  "total_chunks_created": 317777,
+  "total_evaluation_time": 132.6938271522522,
+  "questions_per_second": 21.73424387474267,
+  "timestamp": 1750412729.313979
+}

runs/sentencechunker_1024.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.28120665742024964,
+      "3": 0.42961165048543687,
+      "5": 0.5058945908460472,
+      "10": 0.6119972260748959
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x107c69370>, chunk_size=1024, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x118f49a60>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 811,
+      "3": 1239,
+      "5": 1459,
+      "10": 1765
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.7157663106918335,
+  "chunk_speed_mb_per_sec": 24.284892227680015,
+  "total_chunks_created": 17691,
+  "total_evaluation_time": 49.59530591964722,
+  "questions_per_second": 58.1506645946003,
+  "timestamp": 1750411373.4887922
+}

runs/sentencechunker_128.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.20804438280166435,
+      "3": 0.31345353675450766,
+      "5": 0.35991678224687934,
+      "10": 0.42857142857142855
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x10cd42de0>, chunk_size=128, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10fe30050>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 600,
+      "3": 904,
+      "5": 1038,
+      "10": 1236
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 5.017327284812926,
+  "chunk_speed_mb_per_sec": 13.14486546858272,
+  "total_chunks_created": 162893,
+  "total_evaluation_time": 92.59067273139954,
+  "questions_per_second": 31.14784583503703,
+  "timestamp": 1750411196.5985322
+}

runs/sentencechunker_2048.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.22572815533980584,
+      "3": 0.36927877947295423,
+      "5": 0.4535367545076283,
+      "10": 0.5856449375866851
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x10fe66d20>, chunk_size=2048, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11db5ffe0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 651,
+      "3": 1065,
+      "5": 1308,
+      "10": 1689
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.4737113952636727,
+  "chunk_speed_mb_per_sec": 26.66119107386232,
+  "total_chunks_created": 8780,
+  "total_evaluation_time": 46.66882634162903,
+  "questions_per_second": 61.797140105652176,
+  "timestamp": 1750411423.1875541
+}

runs/sentencechunker_256.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.25208044382801664,
+      "3": 0.4015256588072122,
+      "5": 0.4656726768377254,
+      "10": 0.5565187239944521
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x107e3cf50>, chunk_size=256, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10fe93710>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 727,
+      "3": 1158,
+      "5": 1343,
+      "10": 1605
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.480334115028381,
+  "chunk_speed_mb_per_sec": 18.949931239626835,
+  "total_chunks_created": 75491,
+  "total_evaluation_time": 62.851224184036255,
+  "questions_per_second": 45.88613885316357,
+  "timestamp": 1750411262.276054
+}

runs/sentencechunker_3072.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.19625520110957004,
+      "3": 0.343619972260749,
+      "5": 0.4379334257975035,
+      "10": 0.5894590846047156
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x106f05310>, chunk_size=3072, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x119e693d0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 566,
+      "3": 991,
+      "5": 1263,
+      "10": 1700
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 2.3632036209106446,
+  "chunk_speed_mb_per_sec": 27.907917704231146,
+  "total_chunks_created": 5852,
+  "total_evaluation_time": 44.97272205352783,
+  "questions_per_second": 64.12776163665121,
+  "timestamp": 1750411470.905151
+}

runs/sentencechunker_512.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.30513176144244103,
+      "3": 0.47399445214979197,
+      "5": 0.5547850208044383,
+      "10": 0.6612343966712899
+    }
+  },
+  "metadata": {
+    "chunker_type": "SentenceChunker",
+    "chunker_config": "SentenceChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x1076b1670>, chunk_size=512, chunk_overlap=0, min_sentences_per_chunk=1, min_characters_per_sentence=12, approximate=False, delim=['. ', '! ', '? ', '\\n'], include_delim=prev, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10fe8ffe0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 880,
+      "3": 1367,
+      "5": 1600,
+      "10": 1907
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.0913998365402224,
+  "chunk_speed_mb_per_sec": 21.334054363063697,
+  "total_chunks_created": 36120,
+  "total_evaluation_time": 55.39628601074219,
+  "questions_per_second": 52.061251894048425,
+  "timestamp": 1750411321.090453
+}

runs/tokenchunker_1024_overlap256.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.31137309292649096,
+      "3": 0.47018030513176146,
+      "5": 0.5478502080443828,
+      "10": 0.6726768377253814
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x105f6ae10>, chunk_size=1024, chunk_overlap=256, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11cc3d370>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 898,
+      "3": 1356,
+      "5": 1580,
+      "10": 1940
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.5144939899444565,
+  "chunk_speed_mb_per_sec": 18.765743335858613,
+  "total_chunks_created": 22495,
+  "total_evaluation_time": 64.44851088523865,
+  "questions_per_second": 44.748900484845095,
+  "timestamp": 1750410973.5866199
+}

runs/tokenchunker_128_overlap32.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.19452149791955617,
+      "3": 0.2988904299583911,
+      "5": 0.3564493758668516,
+      "10": 0.4226768377253814
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x107e3c920>, chunk_size=128, chunk_overlap=32, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10b933fe0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 561,
+      "3": 862,
+      "5": 1028,
+      "10": 1219
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.882158160209655,
+  "chunk_speed_mb_per_sec": 16.988512432773632,
+  "total_chunks_created": 179801,
+  "total_evaluation_time": 68.359689950943,
+  "questions_per_second": 42.18860562518125,
+  "timestamp": 1750410763.417346
+}

runs/tokenchunker_2048_overlap512.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.2503467406380028,
+      "3": 0.39563106796116504,
+      "5": 0.481622746185853,
+      "10": 0.5998613037447988
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x1077da210>, chunk_size=2048, chunk_overlap=512, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x10bcbf710>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 722,
+      "3": 1141,
+      "5": 1389,
+      "10": 1730
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.2349274873733513,
+  "chunk_speed_mb_per_sec": 20.387502479774636,
+  "total_chunks_created": 11248,
+  "total_evaluation_time": 61.395374059677124,
+  "questions_per_second": 46.974223126268654,
+  "timestamp": 1750411038.110761
+}

runs/tokenchunker_256_overlap64.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.2721914008321775,
+      "3": 0.42337031900138694,
+      "5": 0.4930651872399445,
+      "10": 0.5901525658807212
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x107665ee0>, chunk_size=256, chunk_overlap=64, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x107e6eab0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 785,
+      "3": 1221,
+      "5": 1422,
+      "10": 1702
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.7745948553085333,
+  "chunk_speed_mb_per_sec": 17.47262810946751,
+  "total_chunks_created": 89919,
+  "total_evaluation_time": 67.27696204185486,
+  "questions_per_second": 42.86757178788459,
+  "timestamp": 1750410836.432965
+}

runs/tokenchunker_3072_overlap768.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.21428571428571427,
+      "3": 0.3602635228848821,
+      "5": 0.4587378640776699,
+      "10": 0.5939667128987517
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x106d19760>, chunk_size=3072, chunk_overlap=768, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x11d458ad0>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 618,
+      "3": 1039,
+      "5": 1323,
+      "10": 1713
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.1229741573333745,
+  "chunk_speed_mb_per_sec": 21.1183598864712,
+  "total_chunks_created": 7504,
+  "total_evaluation_time": 60.06513524055481,
+  "questions_per_second": 48.01454268686603,
+  "timestamp": 1750411101.262261
+}

runs/tokenchunker_512_overlap128.json

@@ -0,0 +1,32 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.332871012482663,
+      "3": 0.4975728155339806,
+      "5": 0.5773231622746186,
+      "10": 0.6792649098474342
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x107e95be0>, chunk_size=512, chunk_overlap=128, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x106ebec00>",
+    "dataset": "gacha",
+    "num_books": 100,
+    "total_questions": 2884,
+    "total_correct": {
+      "1": 960,
+      "3": 1435,
+      "5": 1665,
+      "10": 1959
+    }
+  },
+  "total_corpus_size_mb": 65.95209217071533,
+  "total_time_to_chunk": 3.7282106876373313,
+  "chunk_speed_mb_per_sec": 17.690012098675403,
+  "total_chunks_created": 44963,
+  "total_evaluation_time": 66.33693599700928,
+  "questions_per_second": 43.47502574026063,
+  "timestamp": 1750410905.831187
+}

runs/tokenchunker_A_Christmas_Carol_in_Prose_Bei.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.3103448275862069,
+      "3": 0.4827586206896552,
+      "5": 0.6551724137931034,
+      "10": 0.6896551724137931
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x165d540e0>, chunk_size=512, chunk_overlap=0, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x168fde570>",
+    "total_questions": 29,
+    "total_correct": {
+      "1": 9,
+      "3": 14,
+      "5": 19,
+      "10": 20
+    }
+  },
+  "total_corpus_size_mb": 0.15099716186523438,
+  "total_time_to_chunk": 0.054051876068115234,
+  "chunk_speed_mb_per_sec": 2.793560054695426,
+  "total_chunks_created": 78,
+  "total_evaluation_time": 0.09481215476989746,
+  "questions_per_second": 305.86795617482795,
+  "timestamp": 1750444855.461445
+}

runs/tokenchunker_Daddy_Takes_Us_to_the_Garden.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.27586206896551724,
+      "3": 0.4827586206896552,
+      "5": 0.6206896551724138,
+      "10": 0.8275862068965517
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x118937c20>, chunk_size=512, chunk_overlap=128, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x17c3da960>",
+    "total_questions": 29,
+    "total_correct": {
+      "1": 8,
+      "3": 14,
+      "5": 18,
+      "10": 24
+    }
+  },
+  "total_corpus_size_mb": 0.15877437591552734,
+  "total_time_to_chunk": 0.054865121841430664,
+  "chunk_speed_mb_per_sec": 2.8939036419970363,
+  "total_chunks_created": 112,
+  "total_evaluation_time": 0.12700796127319336,
+  "questions_per_second": 228.33214319235606,
+  "timestamp": 1750444664.238841
+}

runs/tokenchunker_Riders_of_the_Silences.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.2857142857142857,
+      "3": 0.6428571428571429,
+      "5": 0.6428571428571429,
+      "10": 0.8214285714285714
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x1624c76e0>, chunk_size=512, chunk_overlap=0, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x17dec78c0>",
+    "total_questions": 28,
+    "total_correct": {
+      "1": 8,
+      "3": 18,
+      "5": 18,
+      "10": 23
+    }
+  },
+  "total_corpus_size_mb": 0.3561391830444336,
+  "total_time_to_chunk": 0.1116032600402832,
+  "chunk_speed_mb_per_sec": 3.191118099201449,
+  "total_chunks_created": 183,
+  "total_evaluation_time": 0.19881987571716309,
+  "questions_per_second": 140.83099035748418,
+  "timestamp": 1750444826.366462
+}

runs/tokenchunker_The_Reign_of_Greed.json

@@ -0,0 +1,30 @@
+{
+  "metrics": {
+    "recall": {
+      "1": 0.5,
+      "3": 0.6666666666666666,
+      "5": 0.7666666666666667,
+      "10": 0.8333333333333334
+    }
+  },
+  "metadata": {
+    "chunker_type": "TokenChunker",
+    "chunker_config": "TokenChunker(tokenizer=<chonkie.tokenizer.Tokenizer object at 0x141277bc0>, chunk_size=512, chunk_overlap=128, return_type=chunks)",
+    "embedding_model": "Model2VecEmbeddings(model=minishlab/potion-retrieval-32M)",
+    "embedding_model_name": "<model2vec.model.StaticModel object at 0x14324ecf0>",
+    "total_questions": 30,
+    "total_correct": {
+      "1": 15,
+      "3": 20,
+      "5": 23,
+      "10": 25
+    }
+  },
+  "total_corpus_size_mb": 0.6241388320922852,
+  "total_time_to_chunk": 0.22613239288330078,
+  "chunk_speed_mb_per_sec": 2.760059379968539,
+  "total_chunks_created": 412,
+  "total_evaluation_time": 0.4202539920806885,
+  "questions_per_second": 71.38540160313342,
+  "timestamp": 1750444630.732168
+}