Revert to commit 0ebcf8cb99229090371da6c2a35c09a52e3e2bdd

Meta-Llama-3-8B-Instruct-IQ3_M.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4845b035538a703a1c5ac8a362897abb4046b81443939930ec8c7b0323104cdb
-size 3784823360
+oid sha256:f5ece4d5296dc72440d35ccdf221e5db98d44c0e43d703d74de02605fa39093e
+size 3784822976

Meta-Llama-3-8B-Instruct-Q4_K_M.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ba9baf3a7345f705a11878397500fb25174034f0fd784e83aa4a96aaa47735f
-size 4920734272
+oid sha256:ab9e4eec7e80892fd78f74d9a15d0299f1e22121cea44efd68a7a02a3fe9a1da
+size 4920733888

Meta-Llama-3-8B-Instruct-Q5_K_M.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16d824ee771e0e33b762bb3dc3232b972ac8dce4d2d449128fca5081962a1a9e
-size 5732987456
+oid sha256:44a528c93a937c8548e0276ee8df5f8564dcf59bc4aefc66fe6760e396892480
+size 5732987072

Meta-Llama-3-8B-Instruct-Q6_K.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6f1dcba991f8e629531a5f4cf19e4dbc2a89f80fd20737e256b92aac11572f1
-size 6596006464
+oid sha256:13c5c30a3c9404af369a7b66ce1027097ce02a6b5cc0b17a8df5e414c62d93f6
+size 6596006080

Meta-Llama-3-8B-Instruct-Q8_0.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:583c616da14b82930f887f991ab446711da0b029166200b67892d7c9f8f45958
-size 8540770880
+oid sha256:4514087a6e21a05906121ebbe22e5f8610eab5e5db561d3a1be20c7fe43167c8
+size 8540770496

Meta-Llama-3-8B-Instruct.imatrix

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae2eeb4da0dd12a87a816393c5fdb833838be6bb17604fe3918cabc21fa36b43
+size 4988126

README.md

@@ -204,7 +204,6 @@ lm_studio:
 **Model creator:** [meta-llama](https://huggingface.co/meta-llama)<br>
 **Original model**: [Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)<br>
 **GGUF quantization:** provided by [bartowski](https://huggingface.co/bartowski) based on `llama.cpp` PR [6745](https://github.com/ggerganov/llama.cpp/pull/6745)<br>
-**Update April 29 2024:** models have been re-converted and re-quantized using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> commit <a href="https://github.com/ggerganov/llama.cpp/commit/ffe666572f98a686b17a2cd1dbf4c0a982e5ac0a">ffe6665</a>.
 
 ## Model Summary:
 Llama 3 represents a huge update to the Llama family of models. This model is the 8B parameter instruction tuned model, meaning it's small, fast, and tuned for following instructions.<br>