---
datasets:
- wikitext-2-v1
- wikitext
language:
- en
metrics:
- perplexity
- cross_entropy
---

**metrics on 1024 context**:
- valid_perplexity = 14.79
- valid_cross_entropy = 2.69
- train_perplexity = 13.77
- train_cross_entropy = 2.62

**metrics on 252 context**:
- valid_perplexity = 17.35

**metrics on 378 context**:
- valid_perplexity = 16.4

**metrics on 504 context**:
- valid_perplexity = 15.86

**Dependence of the cross entropy loss on the length of the context for prediction**

- x-axis*128 = context length
- y-axis = cross entropy

![image/png](https://cdn-uploads.huggingface.co/production/uploads/63c1ac8cc58fcfeac186bda2/JRsRd01VrzEmTsHySMn0q.png)