Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -12,17 +12,20 @@ tags:
|
|
12 |
- long-context
|
13 |
|
14 |
configs:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
- config_name: core
|
16 |
description: Randomized updates (keys shuffled across key–value pairs). Recommended as the primary/SOTA comparison setting. At the highest stress tier, all tested models (as of Ma 2025) fail to reliably recover the final value.
|
17 |
data_files:
|
18 |
- split: test
|
19 |
path: core.parquet
|
20 |
|
21 |
-
|
22 |
-
description: Non-randomized – clear and strict sequential blocks; prove short context(token=5k-8k) can already have a strong context interference for most LLMs. Even with this well formatted data, many model's the performance still drop rapidly.
|
23 |
-
data_files:
|
24 |
-
- split: test
|
25 |
-
path: additional_sequential.parquet
|
26 |
|
27 |
---
|
28 |
---
|
|
|
12 |
- long-context
|
13 |
|
14 |
configs:
|
15 |
+
|
16 |
+
- config_name: additional_sequential
|
17 |
+
description: Non-randomized – clear and strict sequential blocks; prove short context(token=5k-8k) can already have a strong context interference for most LLMs. Even with this well formatted data, many model's the performance still drop rapidly.
|
18 |
+
data_files:
|
19 |
+
- split: test
|
20 |
+
path: additional_sequential.parquet
|
21 |
+
|
22 |
- config_name: core
|
23 |
description: Randomized updates (keys shuffled across key–value pairs). Recommended as the primary/SOTA comparison setting. At the highest stress tier, all tested models (as of Ma 2025) fail to reliably recover the final value.
|
24 |
data_files:
|
25 |
- split: test
|
26 |
path: core.parquet
|
27 |
|
28 |
+
|
|
|
|
|
|
|
|
|
29 |
|
30 |
---
|
31 |
---
|