Added inference parameter recommendations

README.md

@@ -162,7 +162,8 @@ tokenizer_source: base
 ```
 
 # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
-Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_T145__ZEUS-8B-V2)
+Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_T145__ZEUS-8B-V2)!
+Based on the listed rankings as of 4/12/24, is the top-rank 8B model.
 
 |      Metric       |Value|
 |-------------------|----:|
@@ -172,4 +173,22 @@ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-le
 |MATH Lvl 5 (4-Shot)|21.15|
 |GPQA (0-shot)      | 6.94|
 |MuSR (0-shot)      | 8.24|
-|MMLU-PRO (5-shot)  |32.18|
+|MMLU-PRO (5-shot)  |32.18|
+
+# Inference Settings
+
+Personal recommendations are:
+
+```
+mirostat = 2
+mirostat_eta = 0.1
+mirostat_tau = 4.24
+num_ctx = 4096
+repeat_penalty = 1.4
+temperature = 0.85
+seed = 42
+top_k = 0
+top_p = 0.95
+```
+
+After cross-referencing [this paper on mobile LLMs](https://openreview.net/pdf?id=ahVsd1hy2W) and [this paper on balancing model parameters](https://arxiv.org/html/2408.13586v1).