Update README.md

README.md

@@ -1,59 +1,53 @@
----
-base_model:
-- allura-org/Qwen2.5-32b-RP-Ink
-- Qwen/Qwen2.5-Coder-32B
-- tanliboy/lambda-qwen2.5-32b-dpo-test
-- huihui-ai/QwQ-32B-Preview-abliterated
-- fblgit/TheBeagle-v2beta-32B-MGS
-- Qwen/QwQ-32B
-- Qwen/Qwen2.5-32B-Instruct
-- Qwen/Qwen2.5-32B
-library_name: transformers
-tags:
-- mergekit
-- merge
-
----
-# merge
-
-This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
-
-## Merge Details
+
+# Apollo Model
+
+This is an experimental hybrid reasoning model built on Qwen2.5-32B-Instruct
+
 ### Merge Method
 
 This model was merged using the [Model Stock](https://arxiv.org/abs/2403.19522) merge method using [Qwen/Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct) as a base.
 
-### Models Merged
 
-The following models were included in the merge:
-* [allura-org/Qwen2.5-32b-RP-Ink](https://huggingface.co/allura-org/Qwen2.5-32b-RP-Ink)
-* [Qwen/Qwen2.5-Coder-32B](https://huggingface.co/Qwen/Qwen2.5-Coder-32B)
-* [tanliboy/lambda-qwen2.5-32b-dpo-test](https://huggingface.co/tanliboy/lambda-qwen2.5-32b-dpo-test)
-* [huihui-ai/QwQ-32B-Preview-abliterated](https://huggingface.co/huihui-ai/QwQ-32B-Preview-abliterated)
-* [fblgit/TheBeagle-v2beta-32B-MGS](https://huggingface.co/fblgit/TheBeagle-v2beta-32B-MGS)
-* [Qwen/QwQ-32B](https://huggingface.co/Qwen/QwQ-32B)
-* [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B)
+### Enable reasoning
+
+prompt the LLM with think deeper and step by step
+
+### Example code 
+
+'''
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "rootxhacker/Apollo-v3-32B"
+
+model = AutoModelForCausalLM.from_pretrained(
+model_name,
+torch_dtype="auto",
+device_map="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 
-### Configuration
+prompt = "How many r's are in the word strawberry"
+messages = [
+{"role": "user", "content": prompt}
+]
+text = tokenizer.apply_chat_template(
+messages,
+tokenize=False,
+add_generation_prompt=True
+)
 
-The following YAML configuration was used to produce this model:
+model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
 
-```yaml
-models:
-  - model: Qwen/QwQ-32B #logic
-  - model: huihui-ai/QwQ-32B-Preview-abliterated #uncensored
-  - model: Qwen/Qwen2.5-32B #text generation
-  - model: Qwen/Qwen2.5-32B-Instruct #chat assistant
-  - model: Qwen/Qwen2.5-Coder-32B #coding
-  - model: fblgit/TheBeagle-v2beta-32B-MGS #math
-  - model: tanliboy/lambda-qwen2.5-32b-dpo-test #dpo
-  - model: allura-org/Qwen2.5-32b-RP-Ink # RP
-merge_method: model_stock
-base_model: Qwen/Qwen2.5-32B-Instruct
-normalize: true
-int8_mask: true
-dtype: bfloat16
+generated_ids = model.generate(
+**model_inputs,
+max_new_tokens=32768
+)
+generated_ids = [
+output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+]
 
+response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+print(response)
 
+'''
 
-```