dacorvo HF Staff commited on
Commit
02528ee
·
verified ·
1 Parent(s): 459fa39

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -8672,3 +8672,6 @@ neuronxcc-2.17.194.0+d312836f/MODULE_1e992e5d6bc7377a00f2+165e9558/model.neff fi
8672
  neuronxcc-2.17.194.0+d312836f/MODULE_1e992e5d6bc7377a00f2+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
8673
  neuronxcc-2.17.194.0+d312836f/MODULE_d69c1c06201efa5d12c9+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
8674
  neuronxcc-2.17.194.0+d312836f/MODULE_67d90593bae0dcf21f41+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
8672
  neuronxcc-2.17.194.0+d312836f/MODULE_1e992e5d6bc7377a00f2+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
8673
  neuronxcc-2.17.194.0+d312836f/MODULE_d69c1c06201efa5d12c9+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
8674
  neuronxcc-2.17.194.0+d312836f/MODULE_67d90593bae0dcf21f41+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
8675
+ neuronxcc-2.17.194.0+d312836f/MODULE_044c0ca1270ebc46459d+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
8676
+ neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
8677
+ neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/qwen2/Qwen/Qwen2.5-7B-Instruct/ebf75edb46237363ff9d.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-7B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 18944,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 28,
15
+ "model_type": "qwen2",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 4,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "Qwen/Qwen2.5-7B-Instruct",
24
+ "checkpoint_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
25
+ "continuous_batching": true,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 4,
35
+ "max_context_length": 4096,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 4096,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": false,
43
+ "optimum_neuron_version": "0.3.0.dev2",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "bfloat16",
50
+ "sequence_length": 4096,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "bfloat16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 28,
60
+ "num_hidden_layers": 28,
61
+ "num_key_value_heads": 4,
62
+ "rms_norm_eps": 1e-06,
63
+ "rope_scaling": null,
64
+ "rope_theta": 1000000.0,
65
+ "sliding_window": 131072,
66
+ "tie_word_embeddings": false,
67
+ "use_cache": true,
68
+ "use_sliding_window": false,
69
+ "vocab_size": 152064
70
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_044c0ca1270ebc46459d+431f5505/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_044c0ca1270ebc46459d+431f5505/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_044c0ca1270ebc46459d+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:908243f89c6ebb358d4df109469fad30ab1297ef14aa450841de212ae34f23f5
3
+ size 201967
neuronxcc-2.17.194.0+d312836f/MODULE_044c0ca1270ebc46459d+431f5505/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ee89e570b2cd35fecf7ca90a44e8a27d25c3d3ab01a79d6f220a4cc53f2c3f
3
+ size 6564864
neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44eadc48b2767b96814d84a51afeeaaf50ba1fecd6d7b92ebc057d23448ba607
3
+ size 694958
neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:503267bd9466ca3962afc72296cf4d663dab8aa081c97a72a944a3fdf0551c12
3
+ size 7209984
neuronxcc-2.17.194.0+d312836f/MODULE_ffd03108ca55095db622+165e9558/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa979ed531371d77bfece0f064786b00a1feb9cc0712e5c576b665dfddaa06c3
3
+ size 7404433