Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+llama3-8B-DarkIdol-2.2-Uncensored-1048K.f16.gguf filter=lfs diff=lfs merge=lfs -text
+llama3-8B-DarkIdol-2.2-Uncensored-1048K.q5_k.gguf filter=lfs diff=lfs merge=lfs -text
+llama3-8B-DarkIdol-2.2-Uncensored-1048K.q6_k.gguf filter=lfs diff=lfs merge=lfs -text
+llama3-8B-DarkIdol-2.2-Uncensored-1048K.q8_0.gguf filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+
+---
+license: mit
+language:
+- en
+---
+
+My own (ZeroWw) quantizations.
+output and embed tensors quantized to f16.
+all other tensors quantized to q5_k or q6_k.
+
+Result:
+both f16.q6 and f16.q5 are smaller than q8_0 standard quantization
+and they perform as well as the pure f16.

llama3-8B-DarkIdol-2.2-Uncensored-1048K.f16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce11a9bd5be3980235f7d0522730f86bc44c7e92f143025bea06f1ef877a6437
+size 16068890528

llama3-8B-DarkIdol-2.2-Uncensored-1048K.q5_k.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63fdcb4fd06582dd1270b91619d06ae786664f1c132c3aa66359d60a0a741ab3
+size 7042224032

llama3-8B-DarkIdol-2.2-Uncensored-1048K.q6_k.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61d4ab5653e4110f96a1d64858c5201116f9b1ce68fcc7ba6f04b6aa3b8f7999
+size 7835471776

llama3-8B-DarkIdol-2.2-Uncensored-1048K.q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:698dbf12063c609569e251c44cdac4f19cacb9f6b5de79565cc0d6df667dd974
+size 9525776288