Update README.md

README.md

@@ -1,55 +1,59 @@
 ---
+base_model:
+- tiiuae/falcon-11B
+library_name: transformers
 tags:
-- merge
 - mergekit
+- merge
 - lazymergekit
 - tiiuae/falcon-11B
-base_model:
-- tiiuae/falcon-11B
-- tiiuae/falcon-11B
+license: apache-2.0
+language:
+- cs
 ---
+# sliced
 
-# Falcon2-5.5B-multilingual
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
 
-Falcon2-5.5B-multilingual is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
-* [tiiuae/falcon-11B](https://huggingface.co/tiiuae/falcon-11B)
+## Merge Details
+### Merge Method
+
+This model was pruned using the passthrough merge method.
+
+### Models Merged
+
+The following models were included in the merge:
 * [tiiuae/falcon-11B](https://huggingface.co/tiiuae/falcon-11B)
 
-## 🧩 Configuration
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
 
 ```yaml
 slices:
   - sources:
       - model: tiiuae/falcon-11B
-        layer_range: [0, 24]
+        layer_range: [0, 25]
   - sources:
       - model: tiiuae/falcon-11B
-        layer_range: [55, 59]
+        layer_range: [56, 59]
 merge_method: passthrough
 dtype: bfloat16
 ```
 
-## 💻 Usage
+[PruneMe](https://github.com/arcee-ai/PruneMe) has been utilized using the wikimedia/wikipedia subsets of 10 languages by investigating layer similarity with 2000 samples per language. The layer ranges for pruning were determined based on the averages of each language analysis to maintain performance while reducing model size.
 
-```python
-!pip install -qU transformers accelerate
+![Layer Similarity Plot](https://cdn-uploads.huggingface.co/production/uploads/660c0a02cf274b3ab77dd6b7/A-5Wz8FyUKa0mvqbWdDeu.png)
 
-from transformers import AutoTokenizer
-import transformers
-import torch
+## Direct Use
+Research on large language models; as a foundation for further specialization and finetuning for specific usecases (e.g., summarization, text generation, chatbot, etc.)
 
-model = "ssmits/Falcon2-5.5B-multilingual"
-messages = [{"role": "user", "content": "What is a large language model?"}]
+## Out-of-Scope Use
+Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful.
 
-tokenizer = AutoTokenizer.from_pretrained(model)
-prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-pipeline = transformers.pipeline(
-    "text-generation",
-    model=model,
-    torch_dtype=torch.float16,
-    device_map="auto",
-)
+## Bias, Risks, and Limitations
+Falcon2-5.5B is trained mostly on English, but also German, Spanish, French, Italian, Portuguese, Polish, Dutch, Romanian, Czech, Swedish. It will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online.
 
-outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
-print(outputs[0]["generated_text"])
-```
+## Recommendations
+We recommend users of Falcon2-5.5B to consider finetuning it for the specific set of tasks of interest, and for guardrails and appropriate precautions to be taken for any production use.