Upload README.md with huggingface_hub

README.md

@@ -1,25 +1,18 @@
 ---
 quantized_by: bartowski
 pipeline_tag: text-generation
-language:
-- en
-license_name: nvidia-open-model-license
-tags:
-- nvidia
-- llama-3
-license_link: https://developer.download.nvidia.com/licenses/nvidia-open-model-license-agreement-june-2024.pdf
-license: other
-base_model: nvidia/Llama-3_1-Nemotron-51B-Instruct
 ---
 
 ## Llamacpp imatrix Quantizations of Llama-3_1-Nemotron-51B-Instruct
 
-Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4381">b4381</a> for quantization.
+Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4404">b4404</a> for quantization.
 
 Original model: https://huggingface.co/nvidia/Llama-3_1-Nemotron-51B-Instruct
 
 All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
 
+Run them in [LM Studio](https://lmstudio.ai/)
+
 ## Prompt format
 
 ```
@@ -33,17 +26,22 @@ Today Date: 26 Jul 2024
 {prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
 ```
 
+## What's new:
+
+Fix rope
+
 ## Download a file (not the whole branch) from below:
 
 | Filename | Quant type | File Size | Split | Description |
 | -------- | ---------- | --------- | ----- | ----------- |
-| [Llama-3_1-Nemotron-51B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/tree/main/Llama-3_1-Nemotron-51B-Instruct-Q8_0) | Q8_0 | 54.73GB | true | Extremely high quality, generally unneeded but max available quant. |
+| [Llama-3_1-Nemotron-51B-Instruct-f16.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/tree/main/Llama-3_1-Nemotron-51B-Instruct-f16) | f16 | 103.01GB | true | Full F16 weights. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/tree/main/Llama-3_1-Nemotron-51B-Instruct-Q8_0) | Q8_0 | 54.73GB | true | Extremely high quality, generally unneeded but max available quant. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q6_K_L.gguf) | Q6_K_L | 42.77GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q6_K.gguf) | Q6_K | 42.26GB | false | Very high quality, near perfect, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q5_K_L.gguf) | Q5_K_L | 37.11GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q5_K_M.gguf) | Q5_K_M | 36.47GB | false | High quality, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q5_K_S.gguf) | Q5_K_S | 35.56GB | false | High quality, *recommended*. |
+| [Llama-3_1-Nemotron-51B-Instruct-Q4_1.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q4_1.gguf) | Q4_1 | 32.41GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q4_K_L.gguf) | Q4_K_L | 31.82GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q4_K_M.gguf) | Q4_K_M | 31.04GB | false | Good quality, default size for most use cases, *recommended*. |
 | [Llama-3_1-Nemotron-51B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Llama-3_1-Nemotron-51B-Instruct-GGUF/blob/main/Llama-3_1-Nemotron-51B-Instruct-Q4_K_S.gguf) | Q4_K_S | 29.48GB | false | Slightly lower quality with more space savings, *recommended*. |