Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +22 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/526b4f7540fab77e0309.json +74 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/7d83e791f18e1cf05f15.json +74 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/f4b922640acf4c30ebd7.json +74 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/Qwen/Qwen2.5-0.5B/03662df0c9fbf7869f7a.json +71 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/614847a759b7e4c0d109.json +75 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/77e3fe0deaecdc3d0e6f.json +75 -0
- neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/f75512f31db444015aeb.json +75 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/wrapped_neff.hlo +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.hlo_module.pb +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1649fc77b87fff02e370+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_18642e0fd797db5b7fcb+431f5505/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1b80b788e3a49498f963+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_1df250ef1cf7a7de560f+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_22cf23062ec53b3fd95d+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_24ff9ac2787ce9a1d276+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_284ddd1b388e504631b8+ee23c5ad/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_284ddd1b388e504631b8+ee23c5ad/wrapped_neff.hlo +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_2ef52130792b59d66c66+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_320f2622d4d0c9fdd0f1+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/wrapped_neff.hlo +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_40a0e75a65ac51fdd01a+613edded/model.neff +1 -1
- neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/compile_flags.json +1 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.done +0 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.hlo_module.pb +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.neff +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/wrapped_neff.hlo +3 -0
- neuronxcc-2.18.121.0+9e31e41a/MODULE_4b0ed59c6be672048dc3+431f5505/compile_flags.json +1 -0
.gitattributes
CHANGED
@@ -3130,3 +3130,25 @@ neuronxcc-2.17.194.0+d312836f/MODULE_ec98086d86f184f276f1+165e9558/wrapped_neff.
|
|
3130 |
neuronxcc-2.17.194.0+d312836f/MODULE_01f5b55fd0ce0ced4fe2+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
|
3131 |
neuronxcc-2.17.194.0+d312836f/MODULE_e00ea95e26db168f8dea+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3132 |
neuronxcc-2.17.194.0+d312836f/MODULE_e00ea95e26db168f8dea+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3130 |
neuronxcc-2.17.194.0+d312836f/MODULE_01f5b55fd0ce0ced4fe2+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
|
3131 |
neuronxcc-2.17.194.0+d312836f/MODULE_e00ea95e26db168f8dea+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3132 |
neuronxcc-2.17.194.0+d312836f/MODULE_e00ea95e26db168f8dea+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3133 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3134 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3135 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3136 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3137 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3138 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3139 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3140 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3141 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3142 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_5459a61db398caa4e50a+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
|
3143 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_634439d56d5bb67e9812+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3144 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_634439d56d5bb67e9812+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3145 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_69f077c26bda336334ac+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3146 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_69f077c26bda336334ac+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3147 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_94ba84f59b341a98c066+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3148 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_9aee9a947045c8c3e338+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3149 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_9aee9a947045c8c3e338+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3150 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_9b6ac235ef3c9735ad0c+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3151 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_a08fd8736783f45e5392+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
3152 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_e5f864ff8f5ba3001616+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
|
3153 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_e5f864ff8f5ba3001616+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3154 |
+
neuronxcc-2.18.121.0+9e31e41a/MODULE_f91652bd1c46e6aebbc1+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/526b4f7540fab77e0309.json
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/phi-4-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Phi3ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"auto_map": {},
|
11 |
+
"embd_pdrop": 0.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 16,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 32,
|
16 |
+
"max_position_embeddings": 16384,
|
17 |
+
"model_type": "phi3",
|
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": "yujiepan/phi-4-tiny-random",
|
26 |
+
"checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
|
27 |
+
"continuous_batching": false,
|
28 |
+
"enable_bucketing": false,
|
29 |
+
"ep_degree": 1,
|
30 |
+
"flash_decoding_enabled": false,
|
31 |
+
"fused_qkv": true,
|
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": 100,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 100,
|
42 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": 100,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 0,
|
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": 2,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 1,
|
64 |
+
"original_max_position_embeddings": 16384,
|
65 |
+
"partial_rotary_factor": 1.0,
|
66 |
+
"resid_pdrop": 0.0,
|
67 |
+
"rms_norm_eps": 1e-05,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 250000,
|
70 |
+
"sliding_window": null,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"vocab_size": 100352
|
74 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/7d83e791f18e1cf05f15.json
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/phi-4-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Phi3ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"auto_map": {},
|
11 |
+
"embd_pdrop": 0.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 16,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 32,
|
16 |
+
"max_position_embeddings": 16384,
|
17 |
+
"model_type": "phi3",
|
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": "yujiepan/phi-4-tiny-random",
|
26 |
+
"checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
|
27 |
+
"continuous_batching": false,
|
28 |
+
"enable_bucketing": false,
|
29 |
+
"ep_degree": 1,
|
30 |
+
"flash_decoding_enabled": false,
|
31 |
+
"fused_qkv": true,
|
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": 100,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 100,
|
42 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": "float16",
|
52 |
+
"sequence_length": 100,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 0,
|
55 |
+
"start_rank_id": 0,
|
56 |
+
"target": null,
|
57 |
+
"torch_dtype": "float16",
|
58 |
+
"tp_degree": 2,
|
59 |
+
"vocab_parallel": false
|
60 |
+
},
|
61 |
+
"num_attention_heads": 2,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 1,
|
64 |
+
"original_max_position_embeddings": 16384,
|
65 |
+
"partial_rotary_factor": 1.0,
|
66 |
+
"resid_pdrop": 0.0,
|
67 |
+
"rms_norm_eps": 1e-05,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 250000,
|
70 |
+
"sliding_window": null,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"vocab_size": 100352
|
74 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/phi3/yujiepan/phi-4-tiny-random/f4b922640acf4c30ebd7.json
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/phi-4-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Phi3ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"auto_map": {},
|
11 |
+
"embd_pdrop": 0.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 16,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 32,
|
16 |
+
"max_position_embeddings": 16384,
|
17 |
+
"model_type": "phi3",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 2,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "yujiepan/phi-4-tiny-random",
|
26 |
+
"checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
|
27 |
+
"continuous_batching": true,
|
28 |
+
"enable_bucketing": false,
|
29 |
+
"ep_degree": 1,
|
30 |
+
"flash_decoding_enabled": false,
|
31 |
+
"fused_qkv": true,
|
32 |
+
"glu_mlp": true,
|
33 |
+
"is_chunked_prefill": false,
|
34 |
+
"local_ranks_size": 2,
|
35 |
+
"logical_nc_config": 1,
|
36 |
+
"max_batch_size": 2,
|
37 |
+
"max_context_length": 100,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 100,
|
42 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": false,
|
45 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": "float16",
|
52 |
+
"sequence_length": 100,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 0,
|
55 |
+
"start_rank_id": 0,
|
56 |
+
"target": null,
|
57 |
+
"torch_dtype": "float16",
|
58 |
+
"tp_degree": 2,
|
59 |
+
"vocab_parallel": false
|
60 |
+
},
|
61 |
+
"num_attention_heads": 2,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 1,
|
64 |
+
"original_max_position_embeddings": 16384,
|
65 |
+
"partial_rotary_factor": 1.0,
|
66 |
+
"resid_pdrop": 0.0,
|
67 |
+
"rms_norm_eps": 1e-05,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 250000,
|
70 |
+
"sliding_window": null,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"vocab_size": 100352
|
74 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/Qwen/Qwen2.5-0.5B/03662df0c9fbf7869f7a.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "Qwen/Qwen2.5-0.5B",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Qwen2ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 896,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 4864,
|
13 |
+
"max_position_embeddings": 32768,
|
14 |
+
"max_window_layers": 24,
|
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/Qwen2.5-0.5B",
|
24 |
+
"checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
|
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": "2.18.121.0+9e31e41a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": "float16",
|
50 |
+
"sequence_length": 4096,
|
51 |
+
"sequence_parallel_enabled": false,
|
52 |
+
"speculation_length": 0,
|
53 |
+
"start_rank_id": 0,
|
54 |
+
"target": null,
|
55 |
+
"torch_dtype": "float16",
|
56 |
+
"tp_degree": 2,
|
57 |
+
"vocab_parallel": false
|
58 |
+
},
|
59 |
+
"num_attention_heads": 14,
|
60 |
+
"num_hidden_layers": 24,
|
61 |
+
"num_key_value_heads": 2,
|
62 |
+
"rms_norm_eps": 1e-06,
|
63 |
+
"rope_scaling": null,
|
64 |
+
"rope_theta": 1000000.0,
|
65 |
+
"sliding_window": 32768,
|
66 |
+
"tie_word_embeddings": true,
|
67 |
+
"use_cache": true,
|
68 |
+
"use_mrope": false,
|
69 |
+
"use_sliding_window": false,
|
70 |
+
"vocab_size": 151936
|
71 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/614847a759b7e4c0d109.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/qwen2.5-128k-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Qwen2ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 8,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 16,
|
13 |
+
"max_position_embeddings": 32768,
|
14 |
+
"max_window_layers": 1,
|
15 |
+
"model_type": "qwen2",
|
16 |
+
"neuron": {
|
17 |
+
"_serialized_key": "NxDNeuronConfig",
|
18 |
+
"async_mode": false,
|
19 |
+
"attn_kernel_enabled": false,
|
20 |
+
"batch_size": 1,
|
21 |
+
"capacity_factor": null,
|
22 |
+
"cc_pipeline_tiling_factor": 2,
|
23 |
+
"checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
|
24 |
+
"checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
|
25 |
+
"continuous_batching": false,
|
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": 1,
|
35 |
+
"max_context_length": 100,
|
36 |
+
"max_topk": 256,
|
37 |
+
"mlp_kernel_enabled": false,
|
38 |
+
"mlp_kernel_fuse_residual_add": false,
|
39 |
+
"n_active_tokens": 100,
|
40 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": "float16",
|
50 |
+
"sequence_length": 100,
|
51 |
+
"sequence_parallel_enabled": false,
|
52 |
+
"speculation_length": 0,
|
53 |
+
"start_rank_id": 0,
|
54 |
+
"target": null,
|
55 |
+
"torch_dtype": "float16",
|
56 |
+
"tp_degree": 2,
|
57 |
+
"vocab_parallel": false
|
58 |
+
},
|
59 |
+
"num_attention_heads": 4,
|
60 |
+
"num_hidden_layers": 2,
|
61 |
+
"num_key_value_heads": 2,
|
62 |
+
"rms_norm_eps": 1e-06,
|
63 |
+
"rope_scaling": {
|
64 |
+
"factor": 4.0,
|
65 |
+
"original_max_position_embeddings": 32768,
|
66 |
+
"rope_type": "yarn",
|
67 |
+
"type": "yarn"
|
68 |
+
},
|
69 |
+
"rope_theta": 1000000.0,
|
70 |
+
"sliding_window": 131072,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"use_sliding_window": false,
|
74 |
+
"vocab_size": 152064
|
75 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/77e3fe0deaecdc3d0e6f.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/qwen2.5-128k-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Qwen2ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 8,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 16,
|
13 |
+
"max_position_embeddings": 32768,
|
14 |
+
"max_window_layers": 1,
|
15 |
+
"model_type": "qwen2",
|
16 |
+
"neuron": {
|
17 |
+
"_serialized_key": "NxDNeuronConfig",
|
18 |
+
"async_mode": false,
|
19 |
+
"attn_kernel_enabled": false,
|
20 |
+
"batch_size": 2,
|
21 |
+
"capacity_factor": null,
|
22 |
+
"cc_pipeline_tiling_factor": 2,
|
23 |
+
"checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
|
24 |
+
"checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
|
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": 2,
|
35 |
+
"max_context_length": 100,
|
36 |
+
"max_topk": 256,
|
37 |
+
"mlp_kernel_enabled": false,
|
38 |
+
"mlp_kernel_fuse_residual_add": false,
|
39 |
+
"n_active_tokens": 100,
|
40 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": "float16",
|
50 |
+
"sequence_length": 100,
|
51 |
+
"sequence_parallel_enabled": false,
|
52 |
+
"speculation_length": 0,
|
53 |
+
"start_rank_id": 0,
|
54 |
+
"target": null,
|
55 |
+
"torch_dtype": "float16",
|
56 |
+
"tp_degree": 2,
|
57 |
+
"vocab_parallel": false
|
58 |
+
},
|
59 |
+
"num_attention_heads": 4,
|
60 |
+
"num_hidden_layers": 2,
|
61 |
+
"num_key_value_heads": 2,
|
62 |
+
"rms_norm_eps": 1e-06,
|
63 |
+
"rope_scaling": {
|
64 |
+
"factor": 4.0,
|
65 |
+
"original_max_position_embeddings": 32768,
|
66 |
+
"rope_type": "yarn",
|
67 |
+
"type": "yarn"
|
68 |
+
},
|
69 |
+
"rope_theta": 1000000.0,
|
70 |
+
"sliding_window": 131072,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"use_sliding_window": false,
|
74 |
+
"vocab_size": 152064
|
75 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.3.0.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/f75512f31db444015aeb.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "yujiepan/qwen2.5-128k-tiny-random",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Qwen2ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 8,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 16,
|
13 |
+
"max_position_embeddings": 32768,
|
14 |
+
"max_window_layers": 1,
|
15 |
+
"model_type": "qwen2",
|
16 |
+
"neuron": {
|
17 |
+
"_serialized_key": "NxDNeuronConfig",
|
18 |
+
"async_mode": false,
|
19 |
+
"attn_kernel_enabled": false,
|
20 |
+
"batch_size": 1,
|
21 |
+
"capacity_factor": null,
|
22 |
+
"cc_pipeline_tiling_factor": 2,
|
23 |
+
"checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
|
24 |
+
"checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
|
25 |
+
"continuous_batching": false,
|
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": 1,
|
35 |
+
"max_context_length": 100,
|
36 |
+
"max_topk": 256,
|
37 |
+
"mlp_kernel_enabled": false,
|
38 |
+
"mlp_kernel_fuse_residual_add": false,
|
39 |
+
"n_active_tokens": 100,
|
40 |
+
"neuronxcc_version": "2.18.121.0+9e31e41a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.0.dev1",
|
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": 100,
|
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": 4,
|
60 |
+
"num_hidden_layers": 2,
|
61 |
+
"num_key_value_heads": 2,
|
62 |
+
"rms_norm_eps": 1e-06,
|
63 |
+
"rope_scaling": {
|
64 |
+
"factor": 4.0,
|
65 |
+
"original_max_position_embeddings": 32768,
|
66 |
+
"rope_type": "yarn",
|
67 |
+
"type": "yarn"
|
68 |
+
},
|
69 |
+
"rope_theta": 1000000.0,
|
70 |
+
"sliding_window": 131072,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"use_cache": true,
|
73 |
+
"use_sliding_window": false,
|
74 |
+
"vocab_size": 152064
|
75 |
+
}
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/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-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-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1ed95a9e2c38dbadd3e0a1e10fc4855c1050b806543453e35077e50c740d8351
|
3 |
+
size 88830
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:11e4c2fa5f8cc618cb6178a46af3b5804be0727115c0aaf6532f396ac8865c51
|
3 |
+
size 236544
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_046ebd86c77dc4a94c6c+5be477de/wrapped_neff.hlo
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d66316e616936a42d3d682321b2a7b1e3f4a4f7a8e3dac77e244e457c6da082
|
3 |
+
size 247153
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5db938a12fb905ace1c1dcfb62bcbccf581b6da1f3fa17f5f9db9305aca5069e
|
3 |
+
size 46338
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_0ae7505139b34db6da76+84f3e719/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3b660d33230d51488058809812bad7b6f1e4d5743c533cb3f76fb46f7256ed55
|
3 |
+
size 144384
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.hlo_module.pb
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 80975
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:91c7887234520728a0b4296efbf0dc7438252652c8e6f92a1c81946e04b9e0f7
|
3 |
size 80975
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 226304
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:90b8d13ce16d5915187d63483a95ac66798f2a8ffdce18fad783ff30452cf1e5
|
3 |
size 226304
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1649fc77b87fff02e370+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 134144
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f9447360dc9ced0684cbde35d92a3287a4b7f9b140c793bdcc5240d971ebe976
|
3 |
size 134144
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:12d33afbc0246bedcd17ef20f3faff63f780fba0ac45333cc1a1b1a7e485ca0b
|
3 |
+
size 87437
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1657741ce6205314d0ed+84f3e719/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:175bc2bce1fdfbda183f6a1887e32f793409fbf75e4e04ce6a98814b03201ca3
|
3 |
+
size 246784
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_18642e0fd797db5b7fcb+431f5505/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 103424
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ccd8e60a616162817b781abb8d9324467fe9ef1753a395eebce9fc2aca6fa197
|
3 |
size 103424
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1b80b788e3a49498f963+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 144384
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:519a82ba54d005df23e18cbadf04c9aec44ec960d3596fbcfb7e0a8046ee5ff5
|
3 |
size 144384
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_1df250ef1cf7a7de560f+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 134144
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bbc381718ac0b57d2e6bdd488053d01564eeeec344692713c5befefe33725b5e
|
3 |
size 134144
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_22cf23062ec53b3fd95d+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 144384
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7a46ac7bd4c0fa6e2487dad3e4edf3608f34064f514a513719f6edb30d025b60
|
3 |
size 144384
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_24ff9ac2787ce9a1d276+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 134144
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8c9d23329cbe45857f215901e58bec38621c10d6cf7b7abc30bc49ed573a120f
|
3 |
size 134144
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_284ddd1b388e504631b8+ee23c5ad/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 257024
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ca8e4347a6bcd7003b09e112b2c0b2bbb8bfa27888c1a9f08e29ca8967febde9
|
3 |
size 257024
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_284ddd1b388e504631b8+ee23c5ad/wrapped_neff.hlo
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 268322
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6c3f8980713ccd510d850507aca404deaf37e60db0d726fe2b1f01a18c781bcf
|
3 |
size 268322
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_2ef52130792b59d66c66+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 134144
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35e1d10726f5e488be07bf3e0115705404e5067006659e4d99582da4547173c5
|
3 |
size 134144
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_320f2622d4d0c9fdd0f1+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 144384
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2fe5370194f978f86df6b1be1eacf770dbde11f3962fc49828c597dafbc17b9
|
3 |
size 144384
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff
CHANGED
Binary files a/neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff and b/neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff differ
|
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7dc8b366751b8c761e38be280e38f653f383a4587cd99161f8e6bbf7bf7af8dc
|
3 |
+
size 80121
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_332809dd3646c4e38d3f+84f3e719/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2eea7efae8c93d034963fbe757f33d6f486ed1f7ef483a240c7252ee9a257247
|
3 |
+
size 205824
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/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-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-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a82a31247cb6ba3910f7cf9cbf53cb582bbf0612aa272762726d904f01adbebb
|
3 |
+
size 79134
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5f8e16a72ddcf52590a2d72852188ffca5ab3b2df96be716b98e5681743ede41
|
3 |
+
size 216064
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_3a1bd8b5ecc619e49cdb+5be477de/wrapped_neff.hlo
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa52be530fa4a06c057758daa7c44cb0b7dc3a7174ab4e169baef03b056931c1
|
3 |
+
size 223770
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_40a0e75a65ac51fdd01a+613edded/model.neff
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 144384
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fb0dc5e60f5707655cd02d4dbcf296636479e643f3fbd59b2378163502e5c706
|
3 |
size 144384
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/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-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-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.done
ADDED
File without changes
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:73680a1e5d67702c27dc1ec55e62f35833d4a66ccae1a505b5ef8d608c39a4d3
|
3 |
+
size 46239
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef59413d75d9cdea01f4db5bd7a4f9a6e40d10eab93d8e6d962111fb2c98648e
|
3 |
+
size 2366464
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_49c7199502b64c3c4f30+5be477de/wrapped_neff.hlo
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:92aba45e0270bddb26842459a49f49d8451c11ca03c0cb35d1916be1afb670dc
|
3 |
+
size 2374185
|
neuronxcc-2.18.121.0+9e31e41a/MODULE_4b0ed59c6be672048dc3+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"
|