Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +31 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/c70db4c7be1d872d4380.json +58 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/llamafactory/tiny-random-Llama-3/29c1f48affaff8ffcfac.json +62 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/unsloth/Llama-3.2-1B-Instruct/218ac2136121c43bdfd6.json +63 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama4_text/tiny-random/llama-4/6f976ae1dd1306830cee.json +81 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/mixtral/dacorvo/Mixtral-tiny/3c7054d9dc66bd2bac24.json +58 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/phi3/yujiepan/phi-4-tiny-random/24920d8a9e6a1b9184a3.json +59 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/1e7190a04a0a96096a6e.json +64 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/177ef8d3b7a3fa2f05ba.json +65 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/smollm3/HuggingFaceTB/SmolLM3-3B/ef66b0455dece5e18231.json +134 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_089479b62bff8cacc87f+a02c3a36/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_089479b62bff8cacc87f+a02c3a36/wrapped_neff.hlo +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16c5142e54049280b3a6+24129607/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_3b37aff244c1128985f5+24129607/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_3c3275e8a9f242f022a5+24129607/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_5a2ed470eb752b572dc5+a02c3a36/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5a2ed470eb752b572dc5+a02c3a36/wrapped_neff.hlo +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5a68e035d2f868ab14d1+a02c3a36/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5a68e035d2f868ab14d1+a02c3a36/wrapped_neff.hlo +1 -1
.gitattributes
CHANGED
|
@@ -6136,3 +6136,34 @@ neuronxcc-2.21.33363.0+82129205/MODULE_eb36f973e5afaf932d84+a02c3a36/wrapped_nef
|
|
| 6136 |
neuronxcc-2.21.33363.0+82129205/MODULE_3c3275e8a9f242f022a5+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6137 |
neuronxcc-2.21.33363.0+82129205/MODULE_f20c2d825c9eea3ef8d8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6138 |
neuronxcc-2.21.33363.0+82129205/MODULE_f20c2d825c9eea3ef8d8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6136 |
neuronxcc-2.21.33363.0+82129205/MODULE_3c3275e8a9f242f022a5+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6137 |
neuronxcc-2.21.33363.0+82129205/MODULE_f20c2d825c9eea3ef8d8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6138 |
neuronxcc-2.21.33363.0+82129205/MODULE_f20c2d825c9eea3ef8d8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6139 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6140 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6141 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6142 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6143 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6144 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6145 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6146 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6147 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6148 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6149 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_5a95b2e154e7d24b16d2+80d05c3f/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6150 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_5a95b2e154e7d24b16d2+80d05c3f/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6151 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6436324b782a1664e4df+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6152 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6436324b782a1664e4df+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6153 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6f75af6a30babfcdd019+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6154 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_833bb585fe5b0b130f37+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6155 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_982078e41304be19b706+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6156 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_982078e41304be19b706+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6157 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_a8fc2fc4fb9affdb3f34+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6158 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_aa0c40e8632cc47ef550+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6159 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_d12c75d4a55a1f29a67c+283df001/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6160 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_d4b12c256e167655df02+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6161 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_d4b12c256e167655df02+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6162 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_e43fb6fc70d6ca48c81c+2dde74c7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6163 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_e43fb6fc70d6ca48c81c+2dde74c7/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6164 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_eab0a664432ada643aa8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6165 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_eab0a664432ada643aa8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 6166 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_ebab4b006fd93fdc9f5b+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6167 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_ecf84edc7dafa7cf47bc+6e4949b4/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6168 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_f0c0b96c282628d572b0+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 6169 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_f0c0b96c282628d572b0+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/c70db4c7be1d872d4380.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"GraniteForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"attention_multiplier": 1.0,
|
| 11 |
+
"dtype": "float32",
|
| 12 |
+
"embedding_multiplier": 1.0,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 32,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 64,
|
| 17 |
+
"logits_scaling": 1.0,
|
| 18 |
+
"max_position_embeddings": 2048,
|
| 19 |
+
"mlp_bias": false,
|
| 20 |
+
"model_type": "granite",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 23 |
+
"batch_size": 1,
|
| 24 |
+
"capacity_factor": null,
|
| 25 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
| 26 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
| 27 |
+
"continuous_batching": false,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": true,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 2,
|
| 32 |
+
"max_batch_size": 1,
|
| 33 |
+
"max_context_length": 1024,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 1024,
|
| 36 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"sequence_length": 1024,
|
| 42 |
+
"speculation_length": 0,
|
| 43 |
+
"start_rank_id": 0,
|
| 44 |
+
"target": "trn1",
|
| 45 |
+
"torch_dtype": "float32",
|
| 46 |
+
"tp_degree": 2
|
| 47 |
+
},
|
| 48 |
+
"num_attention_heads": 4,
|
| 49 |
+
"num_hidden_layers": 2,
|
| 50 |
+
"num_key_value_heads": 4,
|
| 51 |
+
"residual_multiplier": 1.0,
|
| 52 |
+
"rms_norm_eps": 1e-06,
|
| 53 |
+
"rope_scaling": null,
|
| 54 |
+
"rope_theta": 10000.0,
|
| 55 |
+
"tie_word_embeddings": false,
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 49152
|
| 58 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/llamafactory/tiny-random-Llama-3/29c1f48affaff8ffcfac.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-Llama-3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 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": "llamafactory/tiny-random-Llama-3",
|
| 24 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"sequence_length": 1024,
|
| 40 |
+
"speculation_length": 0,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "float16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"num_attention_heads": 4,
|
| 47 |
+
"num_hidden_layers": 2,
|
| 48 |
+
"num_key_value_heads": 4,
|
| 49 |
+
"pretraining_tp": 1,
|
| 50 |
+
"rms_norm_eps": 1e-05,
|
| 51 |
+
"rope_scaling": {
|
| 52 |
+
"factor": 8.0,
|
| 53 |
+
"high_freq_factor": 4.0,
|
| 54 |
+
"low_freq_factor": 1.0,
|
| 55 |
+
"original_max_position_embeddings": 8192,
|
| 56 |
+
"rope_type": "llama3"
|
| 57 |
+
},
|
| 58 |
+
"rope_theta": 500000.0,
|
| 59 |
+
"tie_word_embeddings": false,
|
| 60 |
+
"use_cache": true,
|
| 61 |
+
"vocab_size": 128256
|
| 62 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/unsloth/Llama-3.2-1B-Instruct/218ac2136121c43bdfd6.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "2.21.33363.0+82129205",
|
| 35 |
+
"on_device_sampling": false,
|
| 36 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"sequence_length": 4096,
|
| 40 |
+
"speculation_length": 5,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "bfloat16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"num_attention_heads": 32,
|
| 47 |
+
"num_hidden_layers": 16,
|
| 48 |
+
"num_key_value_heads": 8,
|
| 49 |
+
"pretraining_tp": 1,
|
| 50 |
+
"rms_norm_eps": 1e-05,
|
| 51 |
+
"rope_scaling": {
|
| 52 |
+
"factor": 32.0,
|
| 53 |
+
"high_freq_factor": 4.0,
|
| 54 |
+
"low_freq_factor": 1.0,
|
| 55 |
+
"original_max_position_embeddings": 8192,
|
| 56 |
+
"rope_type": "llama3"
|
| 57 |
+
},
|
| 58 |
+
"rope_theta": 500000.0,
|
| 59 |
+
"tie_word_embeddings": true,
|
| 60 |
+
"unsloth_fixed": true,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama4_text/tiny-random/llama-4/6f976ae1dd1306830cee.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "tiny-random/llama-4",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_chunk_size": 128,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"attn_scale": 0.1,
|
| 9 |
+
"attn_temperature_tuning": 4,
|
| 10 |
+
"cache_implementation": "hybrid",
|
| 11 |
+
"dtype": "bfloat16",
|
| 12 |
+
"floor_scale": 8192,
|
| 13 |
+
"for_llm_compressor": false,
|
| 14 |
+
"head_dim": 32,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 32,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"interleave_moe_layer_step": 2,
|
| 19 |
+
"intermediate_size": 64,
|
| 20 |
+
"intermediate_size_mlp": 128,
|
| 21 |
+
"layer_types": [
|
| 22 |
+
"chunked_attention",
|
| 23 |
+
"chunked_attention",
|
| 24 |
+
"chunked_attention",
|
| 25 |
+
"full_attention"
|
| 26 |
+
],
|
| 27 |
+
"max_position_embeddings": 1048576,
|
| 28 |
+
"model_type": "llama4_text",
|
| 29 |
+
"moe_layers": [
|
| 30 |
+
1,
|
| 31 |
+
3
|
| 32 |
+
],
|
| 33 |
+
"neuron": {
|
| 34 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 35 |
+
"batch_size": 1,
|
| 36 |
+
"capacity_factor": null,
|
| 37 |
+
"checkpoint_id": "tiny-random/llama-4",
|
| 38 |
+
"checkpoint_revision": "9e716f5d4d1ffe0a44a15f46f4a12b840439aba4",
|
| 39 |
+
"continuous_batching": false,
|
| 40 |
+
"ep_degree": 1,
|
| 41 |
+
"fused_qkv": false,
|
| 42 |
+
"glu_mlp": true,
|
| 43 |
+
"local_ranks_size": 2,
|
| 44 |
+
"max_batch_size": 1,
|
| 45 |
+
"max_context_length": 1024,
|
| 46 |
+
"max_topk": 256,
|
| 47 |
+
"n_active_tokens": 1024,
|
| 48 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 49 |
+
"on_device_sampling": true,
|
| 50 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 51 |
+
"output_logits": false,
|
| 52 |
+
"pp_degree": 1,
|
| 53 |
+
"sequence_length": 1024,
|
| 54 |
+
"speculation_length": 0,
|
| 55 |
+
"start_rank_id": 0,
|
| 56 |
+
"target": "trn1",
|
| 57 |
+
"torch_dtype": "bfloat16",
|
| 58 |
+
"tp_degree": 2
|
| 59 |
+
},
|
| 60 |
+
"no_rope_layers": [
|
| 61 |
+
1,
|
| 62 |
+
1,
|
| 63 |
+
1,
|
| 64 |
+
0
|
| 65 |
+
],
|
| 66 |
+
"num_attention_heads": 1,
|
| 67 |
+
"num_experts_per_tok": 1,
|
| 68 |
+
"num_hidden_layers": 4,
|
| 69 |
+
"num_key_value_heads": 1,
|
| 70 |
+
"num_local_experts": 8,
|
| 71 |
+
"output_router_logits": false,
|
| 72 |
+
"rms_norm_eps": 1e-05,
|
| 73 |
+
"rope_scaling": null,
|
| 74 |
+
"rope_theta": 500000.0,
|
| 75 |
+
"router_aux_loss_coef": 0.001,
|
| 76 |
+
"router_jitter_noise": 0.0,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"use_cache": true,
|
| 79 |
+
"use_qk_norm": true,
|
| 80 |
+
"vocab_size": 202048
|
| 81 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/mixtral/dacorvo/Mixtral-tiny/3c7054d9dc66bd2bac24.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "dacorvo/Mixtral-tiny",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"MixtralForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "float16",
|
| 10 |
+
"head_dim": 32,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3584,
|
| 15 |
+
"max_position_embeddings": 1024,
|
| 16 |
+
"model_type": "mixtral",
|
| 17 |
+
"neuron": {
|
| 18 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 19 |
+
"batch_size": 1,
|
| 20 |
+
"capacity_factor": null,
|
| 21 |
+
"checkpoint_id": "dacorvo/Mixtral-tiny",
|
| 22 |
+
"checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
|
| 23 |
+
"continuous_batching": false,
|
| 24 |
+
"ep_degree": 1,
|
| 25 |
+
"fused_qkv": false,
|
| 26 |
+
"glu_mlp": true,
|
| 27 |
+
"local_ranks_size": 2,
|
| 28 |
+
"max_batch_size": 1,
|
| 29 |
+
"max_context_length": 1024,
|
| 30 |
+
"max_topk": 256,
|
| 31 |
+
"n_active_tokens": 1024,
|
| 32 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 33 |
+
"on_device_sampling": false,
|
| 34 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 35 |
+
"output_logits": false,
|
| 36 |
+
"pp_degree": 1,
|
| 37 |
+
"sequence_length": 1024,
|
| 38 |
+
"speculation_length": 0,
|
| 39 |
+
"start_rank_id": 0,
|
| 40 |
+
"target": "trn1",
|
| 41 |
+
"torch_dtype": "float16",
|
| 42 |
+
"tp_degree": 2
|
| 43 |
+
},
|
| 44 |
+
"num_attention_heads": 32,
|
| 45 |
+
"num_experts_per_tok": 2,
|
| 46 |
+
"num_hidden_layers": 2,
|
| 47 |
+
"num_key_value_heads": 8,
|
| 48 |
+
"num_local_experts": 8,
|
| 49 |
+
"output_router_logits": false,
|
| 50 |
+
"rms_norm_eps": 1e-05,
|
| 51 |
+
"rope_theta": 10000.0,
|
| 52 |
+
"router_aux_loss_coef": 0.001,
|
| 53 |
+
"router_jitter_noise": 0.0,
|
| 54 |
+
"sliding_window": 4096,
|
| 55 |
+
"tie_word_embeddings": false,
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 32000
|
| 58 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/phi3/yujiepan/phi-4-tiny-random/24920d8a9e6a1b9184a3.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 12 |
+
"embd_pdrop": 0.0,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 16,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 32,
|
| 17 |
+
"max_position_embeddings": 16384,
|
| 18 |
+
"model_type": "phi3",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "yujiepan/phi-4-tiny-random",
|
| 24 |
+
"checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"sequence_length": 1024,
|
| 40 |
+
"speculation_length": 0,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "bfloat16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"num_attention_heads": 2,
|
| 47 |
+
"num_hidden_layers": 2,
|
| 48 |
+
"num_key_value_heads": 1,
|
| 49 |
+
"original_max_position_embeddings": 16384,
|
| 50 |
+
"partial_rotary_factor": 1.0,
|
| 51 |
+
"resid_pdrop": 0.0,
|
| 52 |
+
"rms_norm_eps": 1e-05,
|
| 53 |
+
"rope_scaling": null,
|
| 54 |
+
"rope_theta": 250000,
|
| 55 |
+
"sliding_window": null,
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 100352
|
| 59 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/1e7190a04a0a96096a6e.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 8,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 16,
|
| 14 |
+
"layer_types": [
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention"
|
| 17 |
+
],
|
| 18 |
+
"max_position_embeddings": 32768,
|
| 19 |
+
"max_window_layers": 1,
|
| 20 |
+
"model_type": "qwen2",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 23 |
+
"batch_size": 1,
|
| 24 |
+
"capacity_factor": null,
|
| 25 |
+
"checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
|
| 26 |
+
"checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
|
| 27 |
+
"continuous_batching": false,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": false,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 2,
|
| 32 |
+
"max_batch_size": 1,
|
| 33 |
+
"max_context_length": 1024,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 1024,
|
| 36 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"sequence_length": 1024,
|
| 42 |
+
"speculation_length": 0,
|
| 43 |
+
"start_rank_id": 0,
|
| 44 |
+
"target": "trn1",
|
| 45 |
+
"torch_dtype": "bfloat16",
|
| 46 |
+
"tp_degree": 2
|
| 47 |
+
},
|
| 48 |
+
"num_attention_heads": 4,
|
| 49 |
+
"num_hidden_layers": 2,
|
| 50 |
+
"num_key_value_heads": 2,
|
| 51 |
+
"rms_norm_eps": 1e-06,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 4.0,
|
| 54 |
+
"original_max_position_embeddings": 32768,
|
| 55 |
+
"rope_type": "yarn",
|
| 56 |
+
"type": "yarn"
|
| 57 |
+
},
|
| 58 |
+
"rope_theta": 1000000.0,
|
| 59 |
+
"sliding_window": null,
|
| 60 |
+
"tie_word_embeddings": false,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"use_sliding_window": false,
|
| 63 |
+
"vocab_size": 152064
|
| 64 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/177ef8d3b7a3fa2f05ba.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "optimum-internal-testing/tiny-random-qwen3_moe",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3MoeForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"decoder_sparse_step": 2,
|
| 11 |
+
"dtype": "float32",
|
| 12 |
+
"head_dim": 32,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 64,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 128,
|
| 17 |
+
"max_position_embeddings": 40960,
|
| 18 |
+
"max_window_layers": 1,
|
| 19 |
+
"mlp_only_layers": [],
|
| 20 |
+
"model_type": "qwen3_moe",
|
| 21 |
+
"moe_intermediate_size": 128,
|
| 22 |
+
"neuron": {
|
| 23 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 24 |
+
"batch_size": 1,
|
| 25 |
+
"capacity_factor": null,
|
| 26 |
+
"checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe",
|
| 27 |
+
"checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af",
|
| 28 |
+
"continuous_batching": false,
|
| 29 |
+
"ep_degree": 1,
|
| 30 |
+
"fused_qkv": false,
|
| 31 |
+
"glu_mlp": true,
|
| 32 |
+
"local_ranks_size": 2,
|
| 33 |
+
"max_batch_size": 1,
|
| 34 |
+
"max_context_length": 1024,
|
| 35 |
+
"max_topk": 256,
|
| 36 |
+
"n_active_tokens": 1024,
|
| 37 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 38 |
+
"on_device_sampling": true,
|
| 39 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 40 |
+
"output_logits": false,
|
| 41 |
+
"pp_degree": 1,
|
| 42 |
+
"sequence_length": 1024,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": "trn1",
|
| 46 |
+
"torch_dtype": "float32",
|
| 47 |
+
"tp_degree": 2
|
| 48 |
+
},
|
| 49 |
+
"norm_topk_prob": true,
|
| 50 |
+
"num_attention_heads": 2,
|
| 51 |
+
"num_experts": 8,
|
| 52 |
+
"num_experts_per_tok": 2,
|
| 53 |
+
"num_hidden_layers": 2,
|
| 54 |
+
"num_key_value_heads": 1,
|
| 55 |
+
"output_router_logits": false,
|
| 56 |
+
"rms_norm_eps": 1e-06,
|
| 57 |
+
"rope_scaling": null,
|
| 58 |
+
"rope_theta": 1000000.0,
|
| 59 |
+
"router_aux_loss_coef": 0.001,
|
| 60 |
+
"sliding_window": null,
|
| 61 |
+
"tie_word_embeddings": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"use_sliding_window": false,
|
| 64 |
+
"vocab_size": 151936
|
| 65 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/smollm3/HuggingFaceTB/SmolLM3-3B/ef66b0455dece5e18231.json
ADDED
|
@@ -0,0 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolLM3-3B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"SmolLM3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 65536,
|
| 54 |
+
"max_window_layers": 28,
|
| 55 |
+
"mlp_bias": false,
|
| 56 |
+
"model_type": "smollm3",
|
| 57 |
+
"neuron": {
|
| 58 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 59 |
+
"batch_size": 1,
|
| 60 |
+
"capacity_factor": null,
|
| 61 |
+
"checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
|
| 62 |
+
"checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
|
| 63 |
+
"continuous_batching": false,
|
| 64 |
+
"ep_degree": 1,
|
| 65 |
+
"fused_qkv": true,
|
| 66 |
+
"glu_mlp": true,
|
| 67 |
+
"local_ranks_size": 2,
|
| 68 |
+
"max_batch_size": 1,
|
| 69 |
+
"max_context_length": 1024,
|
| 70 |
+
"max_topk": 256,
|
| 71 |
+
"n_active_tokens": 1024,
|
| 72 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 73 |
+
"on_device_sampling": true,
|
| 74 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 75 |
+
"output_logits": false,
|
| 76 |
+
"pp_degree": 1,
|
| 77 |
+
"sequence_length": 1024,
|
| 78 |
+
"speculation_length": 0,
|
| 79 |
+
"start_rank_id": 0,
|
| 80 |
+
"target": "trn1",
|
| 81 |
+
"torch_dtype": "bfloat16",
|
| 82 |
+
"tp_degree": 2
|
| 83 |
+
},
|
| 84 |
+
"no_rope_layer_interval": 4,
|
| 85 |
+
"no_rope_layers": [
|
| 86 |
+
1,
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
0,
|
| 90 |
+
1,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
0,
|
| 94 |
+
1,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
0,
|
| 98 |
+
1,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
0,
|
| 102 |
+
1,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
0,
|
| 106 |
+
1,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
0,
|
| 110 |
+
1,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
0,
|
| 114 |
+
1,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
0,
|
| 118 |
+
1,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
0
|
| 122 |
+
],
|
| 123 |
+
"num_attention_heads": 16,
|
| 124 |
+
"num_hidden_layers": 36,
|
| 125 |
+
"num_key_value_heads": 4,
|
| 126 |
+
"pretraining_tp": 2,
|
| 127 |
+
"rms_norm_eps": 1e-06,
|
| 128 |
+
"rope_scaling": null,
|
| 129 |
+
"rope_theta": 5000000.0,
|
| 130 |
+
"sliding_window": null,
|
| 131 |
+
"use_cache": false,
|
| 132 |
+
"use_sliding_window": false,
|
| 133 |
+
"vocab_size": 128256
|
| 134 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97f2b999a60c82d86979d0f6b90a6c682dbc9539e64ebe06b9bb34e9f8065e04
|
| 3 |
+
size 84114
|
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6e2652844332eca1427cae990960fd98ac25eb584d577d01490580abbd0b034
|
| 3 |
+
size 246784
|
neuronxcc-2.21.33363.0+82129205/MODULE_04605cea463a0181e7ad+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8771520a6caf1dc15ded687cdf16046be2d5b5d7e47d134e71deb34a171b7214
|
| 3 |
+
size 254967
|
neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2cde5baa98a83cf7d1f88fa3ed4d8421e81f106a689511fb8b313bf33c45b9e
|
| 3 |
+
size 761066
|
neuronxcc-2.21.33363.0+82129205/MODULE_05584a522a6853ba0eb4+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c38cf073637de93c6a4c9bbd5b9773fcb49d5bc61b2e8b9ca9a78f59e54a29c
|
| 3 |
+
size 12493824
|
neuronxcc-2.21.33363.0+82129205/MODULE_089479b62bff8cacc87f+a02c3a36/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6933504
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f21f6bbda57bf7e070cdc973f3fa7c2644644fbf240be8541b7065349c8dcef8
|
| 3 |
size 6933504
|
neuronxcc-2.21.33363.0+82129205/MODULE_089479b62bff8cacc87f+a02c3a36/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7117697
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10393c2cab5611ae66184955d37e5ee955eb6f40016323dcc12ac2c8d4bcc491
|
| 3 |
size 7117697
|
neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07bcaf8a1f374dbd37d11df8fc015c6a3bdfe5ec169e15b6c6167b0561ac31cc
|
| 3 |
+
size 694128
|
neuronxcc-2.21.33363.0+82129205/MODULE_09cd9547033302db1d02+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a294f59751c68c5fc320e02c16ca0a4998a63238dc675c81a4f6328db0682542
|
| 3 |
+
size 625664
|
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a50160f95d4d88b8d0cb2caaeac0d394a6e3c316c2be404ff978c1b367431dfd
|
| 3 |
+
size 82653
|
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:560e291c18ae0e13cd4c74b809ac9c0a69eed02b93b829f21b8bb430cad3976e
|
| 3 |
+
size 216064
|
neuronxcc-2.21.33363.0+82129205/MODULE_127de87388b95258c57e+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ed4c789494bda7c4193add75974fc27cbd033f8f3e6705fba96ed38f97e0714
|
| 3 |
+
size 224275
|
neuronxcc-2.21.33363.0+82129205/MODULE_16c5142e54049280b3a6+24129607/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 71302144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68389367bb30af25de76a4c417e2d0c446b2562faa97b2cbb36aa2a9256dae0d
|
| 3 |
size 71302144
|
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e63fca0f24dc0f69649248d5eac0c2f0cb77dde6f88efae6af8e063e9d3d06da
|
| 3 |
+
size 85024
|
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5beab2ee0ae183b40189a51d8ddd3b6bb09b599e6c594b717a8405394246ee94
|
| 3 |
+
size 277504
|
neuronxcc-2.21.33363.0+82129205/MODULE_23ed678a33869816b749+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:190c8d1353e804bcd4c147b8c311dc08fef8d163541d05d980fd3a4e84492ae3
|
| 3 |
+
size 285717
|
neuronxcc-2.21.33363.0+82129205/MODULE_3b37aff244c1128985f5+24129607/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14654464
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8476d85cba12830b772ba3f157a569bb0c2cb2bbfd65c678b90251ba6f06ef5
|
| 3 |
size 14654464
|
neuronxcc-2.21.33363.0+82129205/MODULE_3c3275e8a9f242f022a5+24129607/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22774784
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4982c2d39eebf629ec91e57f9a85da471cc907955a7ef0ae4523edeb97e47d82
|
| 3 |
size 22774784
|
neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e85d3e2a3ebe4acab2722835a65a1e08488dc52ee22d8f328beef73f4d6686a
|
| 3 |
+
size 83504
|
neuronxcc-2.21.33363.0+82129205/MODULE_4183d5be6001030603fd+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e59b6d51d41360d52fb68989fa190a7fb0d389f7376caa101bfb5012b5bc7c1
|
| 3 |
+
size 328704
|
neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+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-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41cbbba5e64669f88f166e6e00ee0e0e5e2e400e85216c2a34abd2f7cf6b4d57
|
| 3 |
+
size 423848
|
neuronxcc-2.21.33363.0+82129205/MODULE_4386b9df6b20120d67ef+677eeb9d/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac85e4ea5ce014b40fc0e2ef545793aedd869c02f834be33ff3b51e6f9112252
|
| 3 |
+
size 3687424
|
neuronxcc-2.21.33363.0+82129205/MODULE_5a2ed470eb752b572dc5+a02c3a36/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 12411904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb478847dd694f7ad0a83746a6410c240f0f462c807be9ac4197cd6e22395740
|
| 3 |
size 12411904
|
neuronxcc-2.21.33363.0+82129205/MODULE_5a2ed470eb752b572dc5+a02c3a36/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 12558920
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:831569ad3797f3f05f180da2720fe57b55a1be452113afc66b7e4ee63b90a894
|
| 3 |
size 12558920
|
neuronxcc-2.21.33363.0+82129205/MODULE_5a68e035d2f868ab14d1+a02c3a36/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5008384
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d98308e45ddd448ddbc8a54dfab20d673adb007ac2926d05dc406c8b88377fa0
|
| 3 |
size 5008384
|
neuronxcc-2.21.33363.0+82129205/MODULE_5a68e035d2f868ab14d1+a02c3a36/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5192563
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96e00ca7c5c487cc4686abfa7e8f7ea115e132067590de051fbba1fb9db8d0ad
|
| 3 |
size 5192563
|