Commit
•
cb98100
1
Parent(s):
501c1f2
Upload folder using huggingface_hub
Browse files- config.json +3 -3
- llama_float16_tp2_rank0.engine +2 -2
- llama_float16_tp2_rank1.engine +2 -2
- model.cache +0 -0
config.json
CHANGED
@@ -24,7 +24,7 @@
|
|
24 |
"plugin_config": {
|
25 |
"attention_qk_half_accumulation": false,
|
26 |
"bert_attention_plugin": false,
|
27 |
-
"context_fmha_type":
|
28 |
"gemm_plugin": "float16",
|
29 |
"gpt_attention_plugin": "float16",
|
30 |
"identity_plugin": false,
|
@@ -32,14 +32,14 @@
|
|
32 |
"layernorm_quantization_plugin": false,
|
33 |
"lookup_plugin": false,
|
34 |
"nccl_plugin": "float16",
|
35 |
-
"paged_kv_cache":
|
36 |
"quantize_per_token_plugin": false,
|
37 |
"quantize_tensor_plugin": false,
|
38 |
"remove_input_padding": true,
|
39 |
"rmsnorm_plugin": false,
|
40 |
"rmsnorm_quantization_plugin": false,
|
41 |
"smooth_quant_gemm_plugin": false,
|
42 |
-
"tokens_per_block":
|
43 |
"use_custom_all_reduce": false,
|
44 |
"weight_only_groupwise_quant_matmul_plugin": "float16",
|
45 |
"weight_only_quant_matmul_plugin": false
|
|
|
24 |
"plugin_config": {
|
25 |
"attention_qk_half_accumulation": false,
|
26 |
"bert_attention_plugin": false,
|
27 |
+
"context_fmha_type": 0,
|
28 |
"gemm_plugin": "float16",
|
29 |
"gpt_attention_plugin": "float16",
|
30 |
"identity_plugin": false,
|
|
|
32 |
"layernorm_quantization_plugin": false,
|
33 |
"lookup_plugin": false,
|
34 |
"nccl_plugin": "float16",
|
35 |
+
"paged_kv_cache": true,
|
36 |
"quantize_per_token_plugin": false,
|
37 |
"quantize_tensor_plugin": false,
|
38 |
"remove_input_padding": true,
|
39 |
"rmsnorm_plugin": false,
|
40 |
"rmsnorm_quantization_plugin": false,
|
41 |
"smooth_quant_gemm_plugin": false,
|
42 |
+
"tokens_per_block": 64,
|
43 |
"use_custom_all_reduce": false,
|
44 |
"weight_only_groupwise_quant_matmul_plugin": "float16",
|
45 |
"weight_only_quant_matmul_plugin": false
|
llama_float16_tp2_rank0.engine
CHANGED
@@ -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 sha256:2ed39223086dd488a82e72c7fe581965b10db03f196771b3037112f67af58583
|
3 |
+
size 18261922724
|
llama_float16_tp2_rank1.engine
CHANGED
@@ -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 sha256:45bae29bdff0191252a6942f858d707233b59f1c9ce726e3de06e8dabdbf553b
|
3 |
+
size 18261922724
|
model.cache
CHANGED
Binary files a/model.cache and b/model.cache differ
|
|