dacorvo HF Staff commited on
Commit
cc65acb
·
verified ·
1 Parent(s): 9f16c09

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -7403,3 +7403,6 @@ neuronxcc-2.23.6484.0+3b612583/MODULE_e855926ccb3f126d7195+519f203d/model.neff f
7403
  neuronxcc-2.23.6484.0+3b612583/MODULE_ec2c37db809075a6d417+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7404
  neuronxcc-2.23.6484.0+3b612583/MODULE_fd2fd500686b8882cd25+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7405
  neuronxcc-2.23.6484.0+3b612583/MODULE_ff117705b57d7b86f907+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7403
  neuronxcc-2.23.6484.0+3b612583/MODULE_ec2c37db809075a6d417+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7404
  neuronxcc-2.23.6484.0+3b612583/MODULE_fd2fd500686b8882cd25+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7405
  neuronxcc-2.23.6484.0+3b612583/MODULE_ff117705b57d7b86f907+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
7406
+ neuronxcc-2.23.6484.0+3b612583/MODULE_401bf747ba41783a9f3a+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7407
+ neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7408
+ neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.23.6484.0+3b612583/0_REGISTRY/0.4.6.dev3/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/1b359893797dc5a6006d.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
11
+ "head_dim": 64,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 8192,
16
+ "max_position_embeddings": 131072,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
24
+ "checkpoint_revision": null,
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": false,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 4096,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 4096,
34
+ "neuronxcc_version": "2.23.6484.0+3b612583",
35
+ "on_device_sampling": false,
36
+ "optimum_neuron_version": "0.4.6.dev3",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "prefill_chunk_size": 0,
40
+ "sequence_length": 4096,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn1",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 16,
49
+ "num_key_value_heads": 8,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 32.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": true,
61
+ "unsloth_fixed": true,
62
+ "use_cache": true,
63
+ "vocab_size": 128256
64
+ }
neuronxcc-2.23.6484.0+3b612583/MODULE_401bf747ba41783a9f3a+519f203d/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", "--verbose=35", "--logfile=/tmp/nxd_model/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.23.6484.0+3b612583/MODULE_401bf747ba41783a9f3a+519f203d/model.done ADDED
File without changes
neuronxcc-2.23.6484.0+3b612583/MODULE_401bf747ba41783a9f3a+519f203d/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e2a664c844d301a243e62280970d03507de3f9a5726e6a390035f85b3e227a9
3
+ size 1182173
neuronxcc-2.23.6484.0+3b612583/MODULE_401bf747ba41783a9f3a+519f203d/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f569813fff31eac74826a5b12e4febb6405e40615cc7831a8ee5db67d748a06
3
+ size 5970944
neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/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", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/model.done ADDED
File without changes
neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e804daf59ebc68d4a531f1964df53b205e3a3756c12f43a6b1291ffbe9d2c3e6
3
+ size 1368661
neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57b98126f32288d9d8fefb63cde64517744d117f99c0e1f211f5d00a9c3a2dc1
3
+ size 2182144
neuronxcc-2.23.6484.0+3b612583/MODULE_b574571db17ffe273537+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:581fa3e1df8c55a9b0730dfea21d6921dfec137da6e0283ad6c60aecd29adee2
3
+ size 2275060