Update README.md

README.md

@@ -28,7 +28,7 @@ This repository provides large language models developed by the [Research and De
 
 The development was partially supported by [GENIAC](https://www.meti.go.jp/policy/mono_info_service/geniac/index.html).
 
-| Model Variant | 
+| Model Variants | 
 | :--- |
 | [llm-jp-3-1.8b](https://huggingface.co/llm-jp/llm-jp-3-1.8b) |
 | [llm-jp-3-1.8b-instruct](https://huggingface.co/llm-jp/llm-jp-3-1.8b-instruct) |
@@ -82,7 +82,6 @@ print(tokenizer.decode(output))
 |3.7b|28|3072|24|4096|611,844,096|3,171,068,928|
 |13b|40|5120|40|4096|1,019,740,160|12,688,184,320|
 
-
 ## Tokenizer
 
 The tokenizer of this model is based on [huggingface/tokenizers](https://github.com/huggingface/tokenizers) Unigram byte-fallback model.
@@ -133,7 +132,7 @@ The models have been fine-tuned on the following datasets.
 
 ### llm-jp-eval (v1.3.1)
 
-We evaluated using 100 examples from the dev split.
+We evaluated the models using 100 examples from the dev split.
 
 | Model name | average | EL | FA | HE | MC | MR | MT | NLI | QA | RC |
 | :--- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
@@ -172,4 +171,4 @@ llm-jp(at)nii.ac.jp
 
 ## Model Card Authors
 
-Takashi Kodama.
+Takashi Kodama.