Remove redundant files.

Makefile

@@ -37,6 +37,9 @@ llama.cpp/quantize: llama.cpp
 $(MODEL_NAME)-f16.bin: $(HF_FILES) | llama.cpp
 	$(PYTHON) llama.cpp/convert.py --outtype f16 --outfile $@ .
 
+$(MODEL_NAME)-q%_K.bin: $(MODEL_NAME)-q%_K_M.bin
+	ln -s $< $@
+
 $(MODEL_NAME)-q%.bin: $(MODEL_NAME)-f16.bin | llama.cpp/quantize
 	llama.cpp/quantize $< $@ q$*
 

open-llama-7b-q3_K.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:b6f0431fbc6fac5126cd5a855545d9fe27f4a18c6168a99ac37c138a4a75b5c4
-size 3231072384

open-llama-7b-q3_K.bin

@@ -0,0 +1 @@
+open-llama-7b-q3_K_M.bin

open-llama-7b-q4_K.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9a48fc57b63afe6218c2c70ec85c998f88a78d14d009928da7a1f4e8643e73a6
-size 4046946432

open-llama-7b-q4_K.bin

@@ -0,0 +1 @@
+open-llama-7b-q4_K_M.bin

open-llama-7b-q5_K.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:233d50a95afd1f4edb7ba856eb5b7e22fa1cd509a80b08291f4f7bb9cf9a0486
-size 4765483136

open-llama-7b-q5_K.bin

@@ -0,0 +1 @@
+open-llama-7b-q5_K_M.bin