pankajroark commited on
Commit
68b87a7
1 Parent(s): a17d7ca
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ *.engine filter=lfs diff=lfs merge=lfs -text
tp2/config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "builder_config": {
3
+ "apply_query_key_layer_scaling": false,
4
+ "fp8": false,
5
+ "gather_all_token_logits": false,
6
+ "hidden_act": "gelu",
7
+ "hidden_size": 6144,
8
+ "int8": false,
9
+ "max_batch_size": 256,
10
+ "max_input_len": 200,
11
+ "max_num_tokens": null,
12
+ "max_output_len": 200,
13
+ "max_position_embeddings": 8192,
14
+ "multi_query_mode": true,
15
+ "name": "gpt",
16
+ "num_heads": 48,
17
+ "num_kv_heads": 1,
18
+ "num_layers": 40,
19
+ "parallel_build": true,
20
+ "precision": "float16",
21
+ "tensor_parallel": 2,
22
+ "use_parallel_embedding": false,
23
+ "use_prompt_tuning": false,
24
+ "use_refit": false,
25
+ "vocab_size": 49152
26
+ },
27
+ "plugin_config": {
28
+ "attention_qk_half_accumulation": false,
29
+ "bert_attention_plugin": false,
30
+ "context_fmha_type": 1,
31
+ "gemm_plugin": "float16",
32
+ "gpt_attention_plugin": "float16",
33
+ "identity_plugin": false,
34
+ "layernorm_plugin": false,
35
+ "layernorm_quantization_plugin": false,
36
+ "lookup_plugin": false,
37
+ "nccl_plugin": "float16",
38
+ "paged_kv_cache": false,
39
+ "quantize_per_token_plugin": false,
40
+ "quantize_tensor_plugin": false,
41
+ "remove_input_padding": true,
42
+ "rmsnorm_plugin": false,
43
+ "rmsnorm_quantization_plugin": false,
44
+ "smooth_quant_gemm_plugin": false,
45
+ "tokens_per_block": 0,
46
+ "use_custom_all_reduce": false,
47
+ "weight_only_groupwise_quant_matmul_plugin": false,
48
+ "weight_only_quant_matmul_plugin": false
49
+ }
50
+ }
tp2/gpt_float16_tp2_rank0.engine ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6ccb5bf22512ebc9526d73b417b221a6c1d145d3d1eb8be5e4102f3bef0eef1
3
+ size 16243797396
tp2/gpt_float16_tp2_rank1.engine ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c20a53d423f1193decd3165ebbf0338d4e0c6a35bcc7e97bca7848c6573dd265
3
+ size 16243797396