Polish release card typography

README.md

@@ -31,9 +31,9 @@ These math/code results cover **192 cases from a completed 222-case evaluation r
 <details>
 <summary>Test method and exact scope</summary>
 
-Evaluation used the saved step625 adapter with an NF4 backbone, original native prompt IDs and greedy decoding. Math used published GSM8K numeric answers and MathVerify0.9.0; code used the original HumanEval tests in an isolated process. The exact normalized training-question overlap check found zero matches for these frozen evaluation prompts.
+Evaluation used the saved step 625 adapter with an NF4 backbone, original native prompt IDs and greedy decoding. Math used published GSM8K numeric answers and MathVerify 0.9.0; code used the original HumanEval tests in an isolated process. The exact normalized training-question overlap check found zero matches for these frozen evaluation prompts.
 
-The math number above is a format-tolerant secondary analysis. Its checker was chosen after an earlier answer-format artifact and declared before the remaining100 responses were generated. The registered `####` answer-format metric is **111/128 (86.72%)**, retained in evaluation-results.json. Six math responses reached the generation-token limit; none of these math/code cases was cut by the evaluation wall-time limit. Results are from fixed subsets, not full benchmark or leaderboard evaluations.
+The math number above is a format-tolerant secondary analysis. Its checker was chosen after an earlier answer-format artifact and declared before the remaining 100 responses were generated. The registered `####` answer-format metric is **111/128 (86.72%)**, retained in evaluation-results.json. Six math responses reached the generation-token limit; none of these math/code cases was cut by the evaluation wall-time limit. Results are from fixed subsets, not full benchmark or leaderboard evaluations.
 
 </details>
 
@@ -51,7 +51,7 @@ The math number above is a format-tolerant secondary analysis. Its checker was c
 | Examples per optimizer update | 8 |
 | Target modules | Text attention and MLP |
 
-Training used assistant-only labels, complete response reasoning/content, and the original native fine-tuning format. All5,000 examples were consumed exactly once. Vision parameters were frozen. The adapter and native tokenizer are byte-for-byte preserved from the evaluated checkpoint.
+Training used assistant-only labels, complete response reasoning/content, and the original native fine-tuning format. All 5,000 examples were consumed exactly once. Vision parameters were frozen. The adapter and native tokenizer are byte-for-byte preserved from the evaluated checkpoint.
 
 ## Run Bharat
 
@@ -66,8 +66,8 @@ Use `--revision RELEASE_COMMIT_SHA` to pin a public release commit. The helper v
 
 ## Base and data attribution
 
-Base model: [Mistral Small 3.2 24B Instruct](https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506), by **Mistral AI**, revision `95a6d26c4bfb886c58daf9d3f7332c857cb27b43`. Released under Apache2.0; see LICENSE and NOTICE.
+Base model: [Mistral Small 3.2 24B Instruct](https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506), by **Mistral AI**, revision `95a6d26c4bfb886c58daf9d3f7332c857cb27b43`. Released under Apache 2.0; see LICENSE and NOTICE.
 
-Question sources: [NuminaMath-TIR](https://huggingface.co/datasets/AI-MO/NuminaMath-TIR), by Project Numina (Apache2.0), and [OpenCodeInstruct](https://huggingface.co/datasets/nvidia/OpenCodeInstruct), by NVIDIA (CCBY4.0). Only the selected questions were used; the original dataset answers were replaced with newly generated synthetic response targets. See DATA_ATTRIBUTION.md for source credits and license links.
+Question sources: [NuminaMath-TIR](https://huggingface.co/datasets/AI-MO/NuminaMath-TIR), by Project Numina (Apache 2.0), and [OpenCodeInstruct](https://huggingface.co/datasets/nvidia/OpenCodeInstruct), by NVIDIA (CC BY 4.0). Only the selected questions were used; the original dataset answers were replaced with newly generated synthetic response targets. See DATA_ATTRIBUTION.md for source credits and license links.
 
 Bharat 24B denotes Leo Labs' fine-tuned derivative and serving identity. The optional identity system policy supplies that name during ordinary serving. This release provides a model artifact and inference code; it does not send user prompts to a hosted inference API.