dacorvo HF Staff commited on
Commit
60fdc28
·
verified ·
1 Parent(s): a64ce86

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -12119,3 +12119,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_d2e3f4758ceb0fdbb932+747527b0/wrapped_nef
12119
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_1c28f533e563a626d8c5+877608f3/model.neff filter=lfs diff=lfs merge=lfs -text
12120
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d45f3a491b6306d4905c+747527b0/model.neff filter=lfs diff=lfs merge=lfs -text
12121
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d45f3a491b6306d4905c+747527b0/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12119
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_1c28f533e563a626d8c5+877608f3/model.neff filter=lfs diff=lfs merge=lfs -text
12120
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d45f3a491b6306d4905c+747527b0/model.neff filter=lfs diff=lfs merge=lfs -text
12121
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d45f3a491b6306d4905c+747527b0/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12122
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/model.neff filter=lfs diff=lfs merge=lfs -text
12123
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12124
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_dcef693677f812da71c6+877608f3/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/llama/TinyLlama/TinyLlama-1.1B-Chat-v1.0/c11400d965bf3f0f37b9.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
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": 5632,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
23
+ "checkpoint_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 4,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 2048,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 2048,
34
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.1.dev0",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 2048,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn2",
43
+ "torch_dtype": "bfloat16",
44
+ "tp_degree": 4
45
+ },
46
+ "num_attention_heads": 32,
47
+ "num_hidden_layers": 22,
48
+ "num_key_value_heads": 4,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": null,
52
+ "rope_theta": 10000.0,
53
+ "tie_word_embeddings": false,
54
+ "use_cache": true,
55
+ "vocab_size": 32000
56
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47fa61ef848e5b23dcf87c19223200ae0c225e6f41447b9def55eff2529619a8
3
+ size 1529302
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de428267a3eb26be5c8238e61fdf3375690dc68a0e4b90a7a9d927d30936f5b1
3
+ size 1803264
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b9758a1a688ac8ecfcda+747527b0/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8555121f72a894babd08b7a394fa74c2946df2325aed0de16f1b1e03c086cdb8
3
+ size 1905870
neuronxcc-2.21.18209.0+043b1bf7/MODULE_dcef693677f812da71c6+877608f3/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_dcef693677f812da71c6+877608f3/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_dcef693677f812da71c6+877608f3/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72974fb7d35dee5267be77b291a1c8be17f8c7a3b6dedbca59a34e30bfd4f1a3
3
+ size 1692167
neuronxcc-2.21.18209.0+043b1bf7/MODULE_dcef693677f812da71c6+877608f3/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecfcf727962d5441b94d27c086740bfc12db173f436ca456738f7722b09c9c9d
3
+ size 9974784