Quantization made by Richard Erkhov. [Github](https://github.com/RichardErkhov) [Discord](https://discord.gg/pvy7H8DZMG) [Request more models](https://github.com/RichardErkhov/quant_request) A-I-0xtom-7B-slerp - GGUF - Model creator: https://huggingface.co/InnerI/ - Original model: https://huggingface.co/InnerI/A-I-0xtom-7B-slerp/ | Name | Quant method | Size | | ---- | ---- | ---- | | [A-I-0xtom-7B-slerp.Q2_K.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q2_K.gguf) | Q2_K | 2.53GB | | [A-I-0xtom-7B-slerp.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.IQ3_XS.gguf) | IQ3_XS | 2.81GB | | [A-I-0xtom-7B-slerp.IQ3_S.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.IQ3_S.gguf) | IQ3_S | 2.96GB | | [A-I-0xtom-7B-slerp.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q3_K_S.gguf) | Q3_K_S | 2.95GB | | [A-I-0xtom-7B-slerp.IQ3_M.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.IQ3_M.gguf) | IQ3_M | 3.06GB | | [A-I-0xtom-7B-slerp.Q3_K.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q3_K.gguf) | Q3_K | 3.28GB | | [A-I-0xtom-7B-slerp.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q3_K_M.gguf) | Q3_K_M | 3.28GB | | [A-I-0xtom-7B-slerp.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q3_K_L.gguf) | Q3_K_L | 3.56GB | | [A-I-0xtom-7B-slerp.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.IQ4_XS.gguf) | IQ4_XS | 3.67GB | | [A-I-0xtom-7B-slerp.Q4_0.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q4_0.gguf) | Q4_0 | 3.83GB | | [A-I-0xtom-7B-slerp.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.IQ4_NL.gguf) | IQ4_NL | 3.87GB | | [A-I-0xtom-7B-slerp.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q4_K_S.gguf) | Q4_K_S | 3.86GB | | [A-I-0xtom-7B-slerp.Q4_K.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q4_K.gguf) | Q4_K | 4.07GB | | [A-I-0xtom-7B-slerp.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q4_K_M.gguf) | Q4_K_M | 4.07GB | | [A-I-0xtom-7B-slerp.Q4_1.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q4_1.gguf) | Q4_1 | 4.24GB | | [A-I-0xtom-7B-slerp.Q5_0.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q5_0.gguf) | Q5_0 | 4.65GB | | [A-I-0xtom-7B-slerp.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q5_K_S.gguf) | Q5_K_S | 4.65GB | | [A-I-0xtom-7B-slerp.Q5_K.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q5_K.gguf) | Q5_K | 4.78GB | | [A-I-0xtom-7B-slerp.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q5_K_M.gguf) | Q5_K_M | 4.78GB | | [A-I-0xtom-7B-slerp.Q5_1.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q5_1.gguf) | Q5_1 | 5.07GB | | [A-I-0xtom-7B-slerp.Q6_K.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q6_K.gguf) | Q6_K | 5.53GB | | [A-I-0xtom-7B-slerp.Q8_0.gguf](https://huggingface.co/RichardErkhov/InnerI_-_A-I-0xtom-7B-slerp-gguf/blob/main/A-I-0xtom-7B-slerp.Q8_0.gguf) | Q8_0 | 7.17GB | Original model description: --- license: apache-2.0 tags: - merge - mergekit - lazymergekit - 0x0dad0/nous_nous_v2_0 - tomaszki/nous-thirty base_model: - 0x0dad0/nous_nous_v2_0 - tomaszki/nous-thirty model-index: - name: A-I-0xtom-7B-slerp results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 58.19 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 77.64 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 58.74 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 54.78 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 73.24 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 40.18 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=InnerI/A-I-0xtom-7B-slerp name: Open LLM Leaderboard --- # A-I-0xtom-7B-slerp A-I-0xtom-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [0x0dad0/nous_nous_v2_0](https://huggingface.co/0x0dad0/nous_nous_v2_0) * [tomaszki/nous-thirty](https://huggingface.co/tomaszki/nous-thirty) # Avg model loss 0.3912096044793725 I used this testing script that loads your local model, pulls the latest data from cortex and calculates the loss: [avg loss script](https://gist.github.com/romanorac/59ccde7cbf07d8950ef9fb5b5db6a24e) ## 🧩 Configuration ```yaml slices: - sources: - model: 0x0dad0/nous_nous_v2_0 layer_range: [0, 32] - model: tomaszki/nous-thirty layer_range: [0, 32] merge_method: slerp base_model: 0x0dad0/nous_nous_v2_0 parameters: t: - filter: self_attn value: [0, 0.5, 0.3, 0.7, 1] - filter: mlp value: [1, 0.5, 0.7, 0.3, 0] - value: 0.5 dtype: bfloat16 ``` ## 💻 Usage ```python !pip install -qU transformers accelerate from transformers import AutoTokenizer import transformers import torch model = "InnerI/A-I-0xtom-7B-slerp" messages = [{"role": "user", "content": "What is a large language model?"}] 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", ) 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"]) ``` # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_InnerI__A-I-0xtom-7B-slerp) | Metric |Value| |---------------------------------|----:| |Avg. |60.46| |AI2 Reasoning Challenge (25-Shot)|58.19| |HellaSwag (10-Shot) |77.64| |MMLU (5-Shot) |58.74| |TruthfulQA (0-shot) |54.78| |Winogrande (5-shot) |73.24| |GSM8k (5-shot) |40.18|