Update README.md

README.md

@@ -3,8 +3,52 @@ license: apache-2.0
 language:
 - en
 pipeline_tag: text-generation
+tags:
+- bclm
+- pytorch
 ---
 
 <div align="center">
   <img src="https://raw.githubusercontent.com/bcml-labs/bclm-pytorch/refs/heads/main/assets/bclm1.webp" alt="BCLM-1" width="400">
-</div>
+  <p>
+    <b><a href="https://discord.gg/YOUR_INVITE"><b>Discord</b></a> &nbsp;&nbsp;</b>
+    <b><a href="https://github.com/bcml-labs/bclm-pytorch"><b>Architecture</b></a>&nbsp;&nbsp;</b>
+  </p>
+</div>
+
+# BCLM-1-Small
+
+We are excited to release BCLM-1, our first family of flagship language models which deliver compact, predictable performance across a variety of tasks. 
+
+Based on the specialized Bulldog architecture, BCLM-1 is designed to cover wide context lengths while excelling in recall, excelling in tasks such as writing, summarization, and recall.
+
+BCLM-1 provides the foundation for artificial general intelligence (AGI) systems
+
+## BCLM Models
+| Model | Parameters | Variant | Input | Output |
+|---|---|---|---|---|
+| BCLM-1-Small | 123M | Instruction-tuned | Text-only | Text-only |
+| BCLM-1-Medium | 720M | Instruction-tuned | Text-only | Text-only |
+
+## Sample Generations
+BCLM-1 can generate coherent, actionable text in alignment with prompts.
+```
+Photosynthesis is a vital process that occurs in plants, algae, and certain bacteria, allowing them to convert light energy into chemical energy. This process is essential for life on Earth as it provides the primary energy source for nearly all ecosystems.
+```
+```
+Prompt: What are the key components of a successful marketing campaign?
+* A clear and compelling message that resonates with the target audience.
+* Effective use of multiple channels to reach and engage the target audience effectively.
+* Measurable and relevant KPIs to help optimize campaign outcomes.
+```
+
+## Training
+BCLM-1-Small was trained over 24000 steps, achieving a peak training loss of 2.10 (ppl 8). Then, BCLM-1-Small achieved a peak val loss of 1.84 on the SFT dataset.
+
+## Tokenizer & Supported Language
+BCLM-1 uses the `tokenmonster` tokenizer with a vocabulary size of 32000. 
+
+Presently, BCLM-1 only supports English, but this will be expanded in future releases.
+
+## Context Length
+BCLM-1-Small was trained with a max context length of 2048. However, extended context lengths will be included in the next release.