Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +84 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/199a909e91264c5be485.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/4bd6196fdf65916b9e7e.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/da9c5303a7a8399f4728.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/ibm-granite/granite-3.1-2b-instruct/8754088ebe447523faa5.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/1825a9e17d330f6a88b1.json +77 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/2215dc8042463633d3f0.json +77 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/b376023c81bfa4abe720.json +77 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/f251731c5ce664058141.json +77 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/44d1c2cfcc3ebce67ae4.json +78 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/58701e526f964300dd41.json +78 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/b7fe3bbb6c52687bd166.json +78 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/a1ddea966fd8dcd09abe.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/bf9a0ad012138efbed67.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/c0840c79130939a5b5d7.json +73 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/microsoft/Phi-3-mini-4k-instruct/8c3b00bbcd3136816d3f.json +77 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/b4357f21aca6e501d91c.json +74 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/e465629af396fe635f76.json +74 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/e761ab5a54cc018c1cf1.json +74 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/57adbf1ede3e707d0897.json +71 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/70cff495c29999ffef1d.json +71 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/b82afa433d5205813075.json +71 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/193df09e6732fbbb116c.json +75 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/7c0dda1c917359c9b9d9.json +75 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/c2488ca5e3b35076fab4.json +75 -0
- neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen3/Qwen/Qwen3-0.6B/1dec833208e4948af7a0.json +72 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.hlo_module.pb +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.neff +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.hlo_module.pb +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.neff +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.hlo_module.pb +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.neff +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/wrapped_neff.hlo +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.hlo_module.pb +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.neff +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.hlo_module.pb +3 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.neff +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/compile_flags.json +1 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/model.done +0 -0
- neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/model.hlo_module.pb +3 -0
.gitattributes
CHANGED
@@ -10105,3 +10105,87 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_5b8fdb529fb3e11aaec5+ed72d204/model.neff f
|
|
10105 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6eba35a44395e7484ec6+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
|
10106 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cd665ab27cb659881bec+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10107 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cd665ab27cb659881bec+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10105 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6eba35a44395e7484ec6+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
|
10106 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cd665ab27cb659881bec+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10107 |
neuronxcc-2.19.8089.0+8ab9f450/MODULE_cd665ab27cb659881bec+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10108 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10109 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.neff filter=lfs diff=lfs merge=lfs -text
|
10110 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10111 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10112 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10113 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_1b2bdeeb4f1ee675f073+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10114 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_1e73a5b5a7e84e57a4f3+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10115 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_20adb6f2584126fb54d9+c2248236/model.neff filter=lfs diff=lfs merge=lfs -text
|
10116 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_24d3a1081355a38b4dfa+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10117 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_24d3a1081355a38b4dfa+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10118 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_38415f1a88551d65c8cb+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10119 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_44605f8e819b2927ae51+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10120 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_453c6a682d4d9520fb9b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10121 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_453c6a682d4d9520fb9b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10122 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_49a4358be1d6418e0dab+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10123 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_4a80761e06dfb79537d0+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10124 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_4ac4f5ffa7821caa2254+ca355898/model.neff filter=lfs diff=lfs merge=lfs -text
|
10125 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_4ac4f5ffa7821caa2254+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10126 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_52132a4b6612ff8172ff+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10127 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_5292629f22de8a25d49f+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10128 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_5ed208cee0f650dc941b+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10129 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_609761744325080627cc+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10130 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_6108fc3c72e5269b0a19+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10131 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_6108fc3c72e5269b0a19+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10132 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_631cb51f86ae9354545b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10133 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_631cb51f86ae9354545b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10134 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_6642d35addb60ec085b1+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10135 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_6642d35addb60ec085b1+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10136 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_6842d32efdc3bf8cfa8f+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10137 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_7af77f8c231b0912b481+ca355898/model.neff filter=lfs diff=lfs merge=lfs -text
|
10138 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_7af77f8c231b0912b481+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10139 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_7e32ca18bcec0bcbbb95+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10140 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_82e497e3e8402451c963+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10141 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_86263628d1daf779ec21+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10142 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_86263628d1daf779ec21+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10143 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_89d62b3e9349d4bbd3ca+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10144 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_89d62b3e9349d4bbd3ca+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10145 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_8c2a6b120271c37fee40+ca355898/model.neff filter=lfs diff=lfs merge=lfs -text
|
10146 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_8c2a6b120271c37fee40+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10147 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_8d3d2ef7196de2006c25+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10148 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_8d3d2ef7196de2006c25+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10149 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_915f412a8437431eb2b2+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10150 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_915f412a8437431eb2b2+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10151 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_95ed1498cf7e68cf7bb6+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10152 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_a106bf8578a514513a7e+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10153 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_a106bf8578a514513a7e+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10154 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ad5dc858d958ddb16d68+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10155 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ad5dc858d958ddb16d68+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10156 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_aeff7c58d06fb4c123fc+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10157 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b092b55c6af9d765923b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10158 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b092b55c6af9d765923b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10159 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b51d6989e2180ec3eb00+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10160 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b51d6989e2180ec3eb00+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10161 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b5e92b6409d6f348b3ec+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10162 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b5e92b6409d6f348b3ec+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10163 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b75a1633e11cc79ab173+677eeb9d/model.neff filter=lfs diff=lfs merge=lfs -text
|
10164 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b9d2b010664e6041e48d+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10165 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_b9d2b010664e6041e48d+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10166 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_bc97181c1125ffd58a3c+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10167 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_c195f103ee78f704e567+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10168 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_c5100fad76df191ec03f+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10169 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_c7460d110b4eeeb97af0+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10170 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_c839c3566b4b3f8ea4d7+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10171 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_c839c3566b4b3f8ea4d7+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10172 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_cb6febf0237e8ba1d4a2+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10173 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_d6a2494623554e364c4e+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10174 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_d9c5983d78b6c1eaa296+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10175 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_dbcaadcca1c9b6971929+c2248236/model.neff filter=lfs diff=lfs merge=lfs -text
|
10176 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_dcbd37ecd8c47f0d8d55+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10177 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e009dac316c847139d59+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10178 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e009dac316c847139d59+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10179 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e14a323f877d4623a35c+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10180 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e2a6181df5beb66d0222+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10181 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e828a9ab6bbbe87f2110+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10182 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e828a9ab6bbbe87f2110+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10183 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_e9ce27504f605841dd3d+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10184 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ea8966499b21f594b8c4+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10185 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ea8966499b21f594b8c4+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10186 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ef322dd67d1ee094c5f2+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10187 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_f46e2e5cf002555b0f87+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
10188 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_f6fe1f7719e8a4b503de+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
10189 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_fdd3c2470676e1b82f01+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
10190 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_fdd3c2470676e1b82f01+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
10191 |
+
neuronxcc-2.20.9961.0+0acef03a/MODULE_ff351d57db88c064c02a+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/199a909e91264c5be485.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 2,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": true,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 2,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "float16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "float16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/4bd6196fdf65916b9e7e.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 1,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": false,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 1,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "bfloat16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "bfloat16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/da9c5303a7a8399f4728.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 1,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": false,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 1,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "float16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "float16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/granite/ibm-granite/granite-3.1-2b-instruct/8754088ebe447523faa5.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "ibm-granite/granite-3.1-2b-instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"GraniteForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.1,
|
10 |
+
"attention_multiplier": 0.015625,
|
11 |
+
"embedding_multiplier": 12.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 2048,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 8192,
|
16 |
+
"logits_scaling": 8.0,
|
17 |
+
"max_position_embeddings": 131072,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 4,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
28 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
29 |
+
"continuous_batching": true,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 4,
|
39 |
+
"max_context_length": 4096,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 4096,
|
44 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "bfloat16",
|
54 |
+
"sequence_length": 4096,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "bfloat16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 32,
|
64 |
+
"num_hidden_layers": 40,
|
65 |
+
"num_key_value_heads": 8,
|
66 |
+
"residual_multiplier": 0.22,
|
67 |
+
"rms_norm_eps": 1e-05,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 5000000.0,
|
70 |
+
"tie_word_embeddings": true,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49155
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/1825a9e17d330f6a88b1.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 4,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 16,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 64,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 1,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
26 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
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": 131072,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 131072,
|
42 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 131072,
|
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": 4,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 4,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 8.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": false,
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 128256
|
77 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/2215dc8042463633d3f0.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 4,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 16,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 64,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 1,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
26 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 4,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 4,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 8.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": false,
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 128256
|
77 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/b376023c81bfa4abe720.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 4,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 16,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 64,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 1,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
26 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 4,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 4,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 8.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": false,
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 128256
|
77 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/llamafactory/tiny-random-Llama-3/f251731c5ce664058141.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 4,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 16,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 64,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 2,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
26 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 4,
|
62 |
+
"num_hidden_layers": 2,
|
63 |
+
"num_key_value_heads": 4,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 8.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": false,
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 128256
|
77 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/44d1c2cfcc3ebce67ae4.json
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"LlamaForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"head_dim": 64,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 2048,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 8192,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 1,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
26 |
+
"checkpoint_revision": null,
|
27 |
+
"continuous_batching": false,
|
28 |
+
"enable_bucketing": false,
|
29 |
+
"ep_degree": 1,
|
30 |
+
"flash_decoding_enabled": false,
|
31 |
+
"fused_qkv": false,
|
32 |
+
"glu_mlp": true,
|
33 |
+
"is_chunked_prefill": false,
|
34 |
+
"local_ranks_size": 2,
|
35 |
+
"logical_nc_config": 1,
|
36 |
+
"max_batch_size": 1,
|
37 |
+
"max_context_length": 4096,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 4096,
|
42 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": false,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
46 |
+
"output_logits": false,
|
47 |
+
"padding_side": "right",
|
48 |
+
"pp_degree": 1,
|
49 |
+
"qk_layernorm": false,
|
50 |
+
"qkv_kernel_enabled": false,
|
51 |
+
"rpl_reduce_dtype": "bfloat16",
|
52 |
+
"sequence_length": 4096,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 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": 32,
|
62 |
+
"num_hidden_layers": 16,
|
63 |
+
"num_key_value_heads": 8,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 32.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": true,
|
75 |
+
"unsloth_fixed": true,
|
76 |
+
"use_cache": true,
|
77 |
+
"vocab_size": 128256
|
78 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/58701e526f964300dd41.json
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"LlamaForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"head_dim": 64,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 2048,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 8192,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 4,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
26 |
+
"checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
|
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": 4,
|
37 |
+
"max_context_length": 4096,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 4096,
|
42 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 4096,
|
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": 32,
|
62 |
+
"num_hidden_layers": 16,
|
63 |
+
"num_key_value_heads": 8,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 32.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": true,
|
75 |
+
"unsloth_fixed": true,
|
76 |
+
"use_cache": true,
|
77 |
+
"vocab_size": 128256
|
78 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/llama/unsloth/Llama-3.2-1B-Instruct/b7fe3bbb6c52687bd166.json
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"LlamaForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"head_dim": 64,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 2048,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 8192,
|
15 |
+
"max_position_embeddings": 131072,
|
16 |
+
"mlp_bias": false,
|
17 |
+
"model_type": "llama",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 1,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
26 |
+
"checkpoint_revision": null,
|
27 |
+
"continuous_batching": false,
|
28 |
+
"enable_bucketing": false,
|
29 |
+
"ep_degree": 1,
|
30 |
+
"flash_decoding_enabled": false,
|
31 |
+
"fused_qkv": false,
|
32 |
+
"glu_mlp": true,
|
33 |
+
"is_chunked_prefill": false,
|
34 |
+
"local_ranks_size": 2,
|
35 |
+
"logical_nc_config": 1,
|
36 |
+
"max_batch_size": 1,
|
37 |
+
"max_context_length": 4096,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 4096,
|
42 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": false,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
46 |
+
"output_logits": false,
|
47 |
+
"padding_side": "right",
|
48 |
+
"pp_degree": 1,
|
49 |
+
"qk_layernorm": false,
|
50 |
+
"qkv_kernel_enabled": false,
|
51 |
+
"rpl_reduce_dtype": "bfloat16",
|
52 |
+
"sequence_length": 4096,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 5,
|
55 |
+
"start_rank_id": 0,
|
56 |
+
"target": null,
|
57 |
+
"torch_dtype": "bfloat16",
|
58 |
+
"tp_degree": 2,
|
59 |
+
"vocab_parallel": false
|
60 |
+
},
|
61 |
+
"num_attention_heads": 32,
|
62 |
+
"num_hidden_layers": 16,
|
63 |
+
"num_key_value_heads": 8,
|
64 |
+
"pretraining_tp": 1,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_scaling": {
|
67 |
+
"factor": 32.0,
|
68 |
+
"high_freq_factor": 4.0,
|
69 |
+
"low_freq_factor": 1.0,
|
70 |
+
"original_max_position_embeddings": 8192,
|
71 |
+
"rope_type": "llama3"
|
72 |
+
},
|
73 |
+
"rope_theta": 500000.0,
|
74 |
+
"tie_word_embeddings": true,
|
75 |
+
"unsloth_fixed": true,
|
76 |
+
"use_cache": true,
|
77 |
+
"vocab_size": 128256
|
78 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/a1ddea966fd8dcd09abe.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 32,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 3584,
|
14 |
+
"max_position_embeddings": 1024,
|
15 |
+
"model_type": "mixtral",
|
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": "dacorvo/Mixtral-tiny",
|
24 |
+
"checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
|
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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 32,
|
60 |
+
"num_experts_per_tok": 2,
|
61 |
+
"num_hidden_layers": 2,
|
62 |
+
"num_key_value_heads": 8,
|
63 |
+
"num_local_experts": 8,
|
64 |
+
"output_router_logits": false,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_theta": 10000.0,
|
67 |
+
"router_aux_loss_coef": 0.001,
|
68 |
+
"router_jitter_noise": 0.0,
|
69 |
+
"sliding_window": 4096,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 32000
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/bf9a0ad012138efbed67.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 32,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 3584,
|
14 |
+
"max_position_embeddings": 1024,
|
15 |
+
"model_type": "mixtral",
|
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": "dacorvo/Mixtral-tiny",
|
24 |
+
"checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
|
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": 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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 32,
|
60 |
+
"num_experts_per_tok": 2,
|
61 |
+
"num_hidden_layers": 2,
|
62 |
+
"num_key_value_heads": 8,
|
63 |
+
"num_local_experts": 8,
|
64 |
+
"output_router_logits": false,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_theta": 10000.0,
|
67 |
+
"router_aux_loss_coef": 0.001,
|
68 |
+
"router_jitter_noise": 0.0,
|
69 |
+
"sliding_window": 4096,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 32000
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/mixtral/dacorvo/Mixtral-tiny/c0840c79130939a5b5d7.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"head_dim": 32,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 3584,
|
14 |
+
"max_position_embeddings": 1024,
|
15 |
+
"model_type": "mixtral",
|
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": "dacorvo/Mixtral-tiny",
|
24 |
+
"checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
|
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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 32,
|
60 |
+
"num_experts_per_tok": 2,
|
61 |
+
"num_hidden_layers": 2,
|
62 |
+
"num_key_value_heads": 8,
|
63 |
+
"num_local_experts": 8,
|
64 |
+
"output_router_logits": false,
|
65 |
+
"rms_norm_eps": 1e-05,
|
66 |
+
"rope_theta": 10000.0,
|
67 |
+
"router_aux_loss_coef": 0.001,
|
68 |
+
"router_jitter_noise": 0.0,
|
69 |
+
"sliding_window": 4096,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 32000
|
73 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/microsoft/Phi-3-mini-4k-instruct/8c3b00bbcd3136816d3f.json
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "microsoft/Phi-3-mini-4k-instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Phi3ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"auto_map": {
|
11 |
+
"AutoConfig": "microsoft/Phi-3-mini-4k-instruct--configuration_phi3.Phi3Config",
|
12 |
+
"AutoModelForCausalLM": "microsoft/Phi-3-mini-4k-instruct--modeling_phi3.Phi3ForCausalLM"
|
13 |
+
},
|
14 |
+
"embd_pdrop": 0.0,
|
15 |
+
"hidden_act": "silu",
|
16 |
+
"hidden_size": 3072,
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 8192,
|
19 |
+
"max_position_embeddings": 4096,
|
20 |
+
"model_type": "phi3",
|
21 |
+
"neuron": {
|
22 |
+
"_serialized_key": "NxDNeuronConfig",
|
23 |
+
"async_mode": false,
|
24 |
+
"attn_kernel_enabled": false,
|
25 |
+
"batch_size": 4,
|
26 |
+
"capacity_factor": null,
|
27 |
+
"cc_pipeline_tiling_factor": 2,
|
28 |
+
"checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
|
29 |
+
"checkpoint_revision": "0a67737cc96d2554230f90338b163bc6380a2a85",
|
30 |
+
"continuous_batching": true,
|
31 |
+
"enable_bucketing": false,
|
32 |
+
"ep_degree": 1,
|
33 |
+
"flash_decoding_enabled": false,
|
34 |
+
"fused_qkv": true,
|
35 |
+
"glu_mlp": true,
|
36 |
+
"is_chunked_prefill": false,
|
37 |
+
"local_ranks_size": 2,
|
38 |
+
"logical_nc_config": 1,
|
39 |
+
"max_batch_size": 4,
|
40 |
+
"max_context_length": 4096,
|
41 |
+
"max_topk": 256,
|
42 |
+
"mlp_kernel_enabled": false,
|
43 |
+
"mlp_kernel_fuse_residual_add": false,
|
44 |
+
"n_active_tokens": 4096,
|
45 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
46 |
+
"num_cores_per_group": 1,
|
47 |
+
"on_device_sampling": true,
|
48 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
49 |
+
"output_logits": false,
|
50 |
+
"padding_side": "right",
|
51 |
+
"pp_degree": 1,
|
52 |
+
"qk_layernorm": false,
|
53 |
+
"qkv_kernel_enabled": false,
|
54 |
+
"rpl_reduce_dtype": "bfloat16",
|
55 |
+
"sequence_length": 4096,
|
56 |
+
"sequence_parallel_enabled": false,
|
57 |
+
"speculation_length": 0,
|
58 |
+
"start_rank_id": 0,
|
59 |
+
"target": null,
|
60 |
+
"torch_dtype": "bfloat16",
|
61 |
+
"tp_degree": 2,
|
62 |
+
"vocab_parallel": false
|
63 |
+
},
|
64 |
+
"num_attention_heads": 32,
|
65 |
+
"num_hidden_layers": 32,
|
66 |
+
"num_key_value_heads": 32,
|
67 |
+
"original_max_position_embeddings": 4096,
|
68 |
+
"partial_rotary_factor": 1.0,
|
69 |
+
"resid_pdrop": 0.0,
|
70 |
+
"rms_norm_eps": 1e-05,
|
71 |
+
"rope_scaling": null,
|
72 |
+
"rope_theta": 10000.0,
|
73 |
+
"sliding_window": 2047,
|
74 |
+
"tie_word_embeddings": false,
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 32064
|
77 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/b4357f21aca6e501d91c.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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/e465629af396fe635f76.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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/phi3/yujiepan/phi-4-tiny-random/e761ab5a54cc018c1cf1.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": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": true,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/57adbf1ede3e707d0897.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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/70cff495c29999ffef1d.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": 1,
|
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": 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": 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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 4096,
|
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": 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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/Qwen/Qwen2.5-0.5B/b82afa433d5205813075.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": 1,
|
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": 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": 128,
|
36 |
+
"max_topk": 256,
|
37 |
+
"mlp_kernel_enabled": false,
|
38 |
+
"mlp_kernel_fuse_residual_add": false,
|
39 |
+
"n_active_tokens": 128,
|
40 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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": 128,
|
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": 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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/193df09e6732fbbb116c.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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/7c0dda1c917359c9b9d9.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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": true,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen2/yujiepan/qwen2.5-128k-tiny-random/c2488ca5e3b35076fab4.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": "2.20.9961.0+0acef03a",
|
41 |
+
"num_cores_per_group": 1,
|
42 |
+
"on_device_sampling": false,
|
43 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
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-2.20.9961.0+0acef03a/0_REGISTRY/0.3.1.dev0/qwen3/Qwen/Qwen3-0.6B/1dec833208e4948af7a0.json
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "Qwen/Qwen3-0.6B",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"Qwen3ForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"head_dim": 128,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 1024,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 3072,
|
15 |
+
"max_position_embeddings": 40960,
|
16 |
+
"max_window_layers": 28,
|
17 |
+
"model_type": "qwen3",
|
18 |
+
"neuron": {
|
19 |
+
"_serialized_key": "NxDNeuronConfig",
|
20 |
+
"async_mode": false,
|
21 |
+
"attn_kernel_enabled": false,
|
22 |
+
"batch_size": 4,
|
23 |
+
"capacity_factor": null,
|
24 |
+
"cc_pipeline_tiling_factor": 2,
|
25 |
+
"checkpoint_id": "Qwen/Qwen3-0.6B",
|
26 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
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": 4,
|
37 |
+
"max_context_length": 4096,
|
38 |
+
"max_topk": 256,
|
39 |
+
"mlp_kernel_enabled": false,
|
40 |
+
"mlp_kernel_fuse_residual_add": false,
|
41 |
+
"n_active_tokens": 4096,
|
42 |
+
"neuronxcc_version": "2.20.9961.0+0acef03a",
|
43 |
+
"num_cores_per_group": 1,
|
44 |
+
"on_device_sampling": false,
|
45 |
+
"optimum_neuron_version": "0.3.1.dev0",
|
46 |
+
"output_logits": false,
|
47 |
+
"padding_side": "right",
|
48 |
+
"pp_degree": 1,
|
49 |
+
"qk_layernorm": false,
|
50 |
+
"qkv_kernel_enabled": false,
|
51 |
+
"rpl_reduce_dtype": "bfloat16",
|
52 |
+
"sequence_length": 4096,
|
53 |
+
"sequence_parallel_enabled": false,
|
54 |
+
"speculation_length": 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": 16,
|
62 |
+
"num_hidden_layers": 28,
|
63 |
+
"num_key_value_heads": 8,
|
64 |
+
"rms_norm_eps": 1e-06,
|
65 |
+
"rope_scaling": null,
|
66 |
+
"rope_theta": 1000000,
|
67 |
+
"sliding_window": null,
|
68 |
+
"tie_word_embeddings": true,
|
69 |
+
"use_cache": true,
|
70 |
+
"use_sliding_window": false,
|
71 |
+
"vocab_size": 151936
|
72 |
+
}
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2407e2272d3d144dbf410618284b0f2c0f2663b358d5958bc4343a79b1717945
|
3 |
+
size 81899
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_05fa2bbe33c465fa6028+ed72d204/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:27eaaca8b2bcbc0ad2b82118de895ed5ef25f6aef54272664ad56dcece9fea5d
|
3 |
+
size 236544
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09bc195566c41e9d9f79bcb7ff575d29e270b3f183325ed7b9cb5c5c05369605
|
3 |
+
size 82456
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0d1769ff0e8722ea7bd3+c2248236/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:06b904f1a0e5ca3bbf73c9e84e696fa4d2f2c7bff8e8e23caef0b01b4ba92330
|
3 |
+
size 420864
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6dc564c78c9e2a9c67927b13cfeb427f1755a039f1efbb3fc65845e0b6ac57df
|
3 |
+
size 84609
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:63cf5899873bccda25078727cddbe0a964348d337fd6ca871dbb54257bd8fb7b
|
3 |
+
size 185344
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_0f186c2f1fd5ea8194e6+a9d440f5/wrapped_neff.hlo
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7ee7cadb30ac2bd4f7d0b08a8c7c3490e4d801cce2b8aecbb7d21eae6e3e9eaf
|
3 |
+
size 193114
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a88073b5193e8858262325e256514fbe34bda1acf82c1d544870c9486ed9584d
|
3 |
+
size 430694
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_12a77fbb68e4df67e3ed+ed72d204/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef627b07ecba743bba6d5de0418836f03275e668589e944635437471b4bc8ad2
|
3 |
+
size 31120384
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options='--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --verify-hlo=false' --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:69ec8adc0e62d42819e2e31a593d500e4ecc872b5b66fad58c6413dcecc02964
|
3 |
+
size 10362
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_147e4bf1152204dd48de+ae6a382b/model.neff
ADDED
Binary file (42 kB). View file
|
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options='--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --verify-hlo=false' --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/model.done
ADDED
File without changes
|
neuronxcc-2.20.9961.0+0acef03a/MODULE_159ba669f1f1a100fca9+ae6a382b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9301197ab622d6cf7fc9b997b77ddb90e143075decfbd981455569f9c3a19328
|
3 |
+
size 7011
|