l3utterfly
commited on
Commit
β’
16a446f
1
Parent(s):
1e7e281
Add PTE files for context sizes 2048, 4096, 8192
Browse files- suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx2048.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx2048.pte +2 -2
- suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx4096.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx4096.pte +2 -2
- suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx8192.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx8192.pte +2 -2
suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx2048.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx2048.pte
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3571a0f94164ffa51a804fd433a6434ee85ccd12f15f8dbee7beafc2a8be6640
|
3 |
+
size 4169560736
|
suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx4096.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx4096.pte
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dbd5a123d4f1b0ee79fee27faab634fe6f5a78c8d8e99a61047d8feb57e880c9
|
3 |
+
size 4171657888
|
suzume-llama-3-8B-multilingual_kv_sdpa_xnn_qe_4_32_ctx8192.pte β suzume-llama-3-8B-multilingual_kv2_sdpa_xnn_qe_4_32_ctx8192.pte
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eff0f7df355ba65b329bc35ba849fb21644c2661c15e4ebae87956f4b99bc11e
|
3 |
+
size 4175852192
|