optimum-internal-testing-user commited on
Commit
328694e
·
verified ·
1 Parent(s): 81cf689

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -3632,3 +3632,7 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_85a3b00cb47dd0b8aebf+ed72d204/model.neff f
3632
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9430ff6a35fc239154d1+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3633
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fdd3c2470676e1b82f01+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
3634
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fdd3c2470676e1b82f01+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
3632
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9430ff6a35fc239154d1+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3633
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fdd3c2470676e1b82f01+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
3634
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fdd3c2470676e1b82f01+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3635
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
3636
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3637
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_b75a1633e11cc79ab173+677eeb9d/model.neff filter=lfs diff=lfs merge=lfs -text
3638
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_cb6ef7b1205d1d19bf46+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0.dev4/llama/unsloth/Llama-3.2-1B-Instruct/2cebcd33ffcf970fd698.json ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.2-1B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
26
+ "checkpoint_revision": null,
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": false,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 4096,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 4096,
42
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.3.0.dev4",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 4096,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 5,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 32,
62
+ "num_hidden_layers": 16,
63
+ "num_key_value_heads": 8,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 32.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": true,
75
+ "unsloth_fixed": true,
76
+ "use_cache": true,
77
+ "vocab_size": 128256
78
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cfceaa4c3d4cd40b71d3e19d1a5214c3763ccd682e664ab43269d5ea18bbd8a
3
+ size 379979
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1176c181d7e8d34df3cad9dddd37b4e79f21ad685a41894906b20ec944e7396c
3
+ size 1434624
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6642d35addb60ec085b1+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e786fcf2924685e9ef035aea47c76b68ffe7735819dd70eaa354cc174efa241
3
+ size 1521594
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b75a1633e11cc79ab173+677eeb9d/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/speculation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b75a1633e11cc79ab173+677eeb9d/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b75a1633e11cc79ab173+677eeb9d/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9958c7d5f720b797c18b96c49efd066e29ec798196fccc80c344806841aea66d
3
+ size 415162
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b75a1633e11cc79ab173+677eeb9d/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4726b763d462ea864d517dfa56c58e0c3a3293e029d304a658092b12e093bac0
3
+ size 2694144
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cb6ef7b1205d1d19bf46+ed72d204/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cb6ef7b1205d1d19bf46+ed72d204/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cb6ef7b1205d1d19bf46+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4949c3d5a68f0516e9b23e07f97a659a019e4604259d3644eb220ca01c8e881
3
+ size 387278
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cb6ef7b1205d1d19bf46+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a199bdfce15b814e14fbf1ac94f9423835204f84fd606350533476cfa7d5fecf
3
+ size 31878144