Update README.md

README.md

@@ -32,7 +32,7 @@ The **QA Increasing Context Length** dataset is designed to facilitate benchmark
 
 * **Key features**
 
-  1. A single CSV (`longbench_all_buckets_100.csv`) containing examples from five context‐length buckets: **2 K**, **4 K**, **8 K**, **16 K**, and **32 K** tokens.
+  1. A single CSV (`longbench_all_buckets_100.csv`) containing examples from five context‐length buckets: **3 K**, **4 K**, **8 K**, **16 K**, and **32 K** tokens.
   2. Each row includes a complete (potentially multi‐paragraph) passage, a target question, and its ground‐truth answer, along with metadata fields that facilitate grouping, filtering, or statistical analysis.
   3. Examples are drawn from diverse domains (scientific articles, technical reports, web pages, etc.), as indicated by the `dataset` field.
 
@@ -40,7 +40,7 @@ The **QA Increasing Context Length** dataset is designed to facilitate benchmark
 
 * **File format**: Comma‐separated values (UTF-8 encoded)
 * **Number of rows**: Varies by bucket (typically 100 examples per bucket)
-* **Context lengths**: 5 (`“2k”`, `“4k”`, `“8k”`, `“16k”`, `“32k”`)
+* **Context lengths**: 5 (`“3k”`, `“4k”`, `“8k”`, `“16k”`, `“32k”`)
 
 ### 2.1. Column Descriptions
 
@@ -48,7 +48,7 @@ Each row (example) has six columns:
 
 | Column Name        | Type     | Description                                                                                                |
 | ------------------ | -------- | ---------------------------------------------------------------------------------------------------------- |
-| **context**        | `string` | A (long) text passage whose token count falls into one of the predefined buckets (2 K – 32 K).             |
+| **context**        | `string` | A (long) text passage whose token count falls into one of the predefined buckets (3 K – 32 K).             |
 | **question**       | `string` | A natural‐language question referring to information contained in `context`.                               |
 | **answer**         | `string` | The ground‐truth answer (text span or summary) extracted from the context.                                 |
 | **length**         | `int`    | The exact token count of the `context` (as measured by a standard tokenizer, e.g., T5/BPE).                |
@@ -57,15 +57,15 @@ Each row (example) has six columns:
 
 * **Context buckets (`context_range`)**
 
-  * `"2k"`:  1 500 – 2 499 tokens (approximate; exact boundaries may vary)
+  * `"3k"`:  1 500 – 3 000 tokens (approximate; exact boundaries may vary)
 
-  * `"4k"`:  3 000 – 4 999 tokens
+  * `"4k"`:  3 000 – 3 999 tokens
 
-  * `"8k"`:  6 000 – 9 999 tokens
+  * `"8k"`:  4 000 – 7 999 tokens
 
-  * `"16k"`: 12 000 – 17 999 tokens
+  * `"16k"`: 8 000 – 15 999 tokens
 
-  * `"32k"`: 24 000 – 34 999 tokens
+  * `"32k"`: 16 000 – 31 999 tokens
 
   > **Note**: The buckets are chosen to stress‐test long‐context inference. The exact cutoff may be implementation‐dependent, but each row’s `length` field indicates the precise token count.
 
@@ -92,7 +92,7 @@ By default, the single CSV is assigned a “train” split. You can then filter
 # Example: count examples per bucket
 from collections import Counter
 counts = Counter(dataset["train"]["context_range"])
-print(counts)  # e.g., {'2k': 100, '4k': 100, '8k': 100, '16k': 100, '32k': 100}
+print(counts)  # e.g., {'3k': 100, '4k': 100, '8k': 100, '16k': 100, '32k': 100}
 
 # Filter to only 16k‐token contexts
 ds_16k = dataset["train"].filter(lambda x: x["context_range"] == "16k")