Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +206 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/03b0c107d1cede36875199a5d51decfe04c473de2af9999f8577a028d74d0ab4/ec1560a24138c907085c.json +189 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/04e22cfac37c408caa51d61b493ee185fc351c2f9db86171e024b5d0ad5b25c1/c56a8bce0577ce31a594.json +62 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/0cc25526e2cfc37a8875a3752f33c4d7505d8a07b869d0f3f41915cf6e763b74/2132439b615164bef4eb.json +59 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/0cc25526e2cfc37a8875a3752f33c4d7505d8a07b869d0f3f41915cf6e763b74/da74b3645f29f4218169.json +59 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/6ed6aa77ee480d24287c.json +82 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/748542be1bcc5fe3cdf1.json +82 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/9a130a0e2b7177d1dea1.json +82 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/196ef14bd9a3e0d3035c.json +83 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/3474ddeda6545f7e5db7.json +83 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/0d1cbffb3e66b482c261.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/61ff8eae2078e66cabfd.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/24892d985c7b4406e3d9.json +63 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/dedf7a5ae53bb6e1b3fc.json +63 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/df8b960809b8e9d97bd2.json +63 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/73707b485eab9008c7aba7f5dad0ce2384ac685318d5f888c12fa0d81ed90b19/309698d1c71c3e0c8eb8.json +135 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/73707b485eab9008c7aba7f5dad0ce2384ac685318d5f888c12fa0d81ed90b19/8d67a5db932b99602641.json +135 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/0b70cb2cd9d99c1bdf34.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/805a8935ea4fbf9c4df6.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/fd610699b7c75d5fa5cc.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/5df183a9a537186b793f.json +169 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/73e765061ca0edc0984c.json +125 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/c6310bffbfbf29e5d927.json +124 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/190c48cba7eeccd29748.json +64 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/4709e472b680eafc6cc9.json +64 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/552e8cb4a475fa6bf79e.json +64 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/d391df1950611ad2a02d.json +64 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/e0a07fcbaaaca6e450fabdb638553525f4bfb07bcac85320ca02556f31cd98cf/68b8b60ad74d23a1ba59.json +130 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3/unsloth/gemma-3-4b-it/68b8b60ad74d23a1ba59.json +130 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3_text/unsloth/gemma-3-270m-it/748542be1bcc5fe3cdf1.json +82 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/granite/ibm-granite/granite-3.1-2b-instruct/2132439b615164bef4eb.json +59 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/idefics3/HuggingFaceTB/SmolVLM-256M-Instruct/ec1560a24138c907085c.json +189 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/llamafactory/tiny-random-Llama-3/24892d985c7b4406e3d9.json +63 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/unsloth/Llama-3.2-1B-Instruct/552e8cb4a475fa6bf79e.json +64 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen2/Qwen/Qwen2.5-0.5B/3474ddeda6545f7e5db7.json +83 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-0.6B/0b70cb2cd9d99c1bdf34.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-Embedding-0.6B/61ff8eae2078e66cabfd.json +88 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/llamafactory/tiny-random-qwen3/c56a8bce0577ce31a594.json +62 -0
- neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/smollm3/HuggingFaceTB/SmolLM3-3B/8d67a5db932b99602641.json +135 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/compile_flags.json +1 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.done +0 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.hlo_module.pb +3 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.neff +3 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/compile_flags.json +1 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.done +0 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.hlo_module.pb +3 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.neff +3 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/wrapped_neff.hlo +3 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0632e5fdcca8738ae989+12dd9119/compile_flags.json +1 -0
- neuronxcc-2.26.6360.0+6f180f47/MODULE_0632e5fdcca8738ae989+12dd9119/model.done +0 -0
.gitattributes
CHANGED
|
@@ -17565,3 +17565,209 @@ neuronxcc-2.24.5133.0+58f8de22/MODULE_5983480387230594376+bad9cf09/model.neff fi
|
|
| 17565 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_7343260162999572602+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17566 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_7903529212460185893+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17567 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_8173596499745875930+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17565 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_7343260162999572602+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17566 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_7903529212460185893+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17567 |
neuronxcc-2.24.5133.0+58f8de22/MODULE_8173596499745875930+bad9cf09/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17568 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17569 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17570 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17571 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0632e5fdcca8738ae989+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17572 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_06618d6c5e8bd5e409c0+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17573 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_073e3ae283183bf46f7e+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17574 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_098c7b2bb282dd07336b+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17575 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0a0e90258cde52021d6c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17576 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0a0e90258cde52021d6c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17577 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0ad611735b0e10f907be+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17578 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0ad611735b0e10f907be+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17579 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0b4c7dcdd85918a819c3+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17580 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0b4c7dcdd85918a819c3+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17581 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0b6b86331e957bacee2d+1d5aa319/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17582 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0b6b86331e957bacee2d+1d5aa319/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17583 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_0f5a9802a78ccacfe236+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17584 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_1411a1bcc1201278ff08+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17585 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_147a9ca8d0079fe60a3f+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17586 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_147a9ca8d0079fe60a3f+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17587 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_165ddf347e87727e1e3f+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17588 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_17bdfe7645d8e87128ff+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17589 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_186a5c4840fd4d6f90d4+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17590 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_19787c71cd7be643f3f9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17591 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_19787c71cd7be643f3f9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17592 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_1eb086c415411ddf9914+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17593 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_1eb086c415411ddf9914+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17594 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_21dad3fc72fb243184b9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17595 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_21dad3fc72fb243184b9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17596 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_24186092e38c4b24437c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17597 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_266d2d16441fb5da6c03+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17598 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_28f077d239104f60dc85+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17599 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_28f077d239104f60dc85+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17600 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_293aaa3b245c1b274ddc+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17601 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_293aaa3b245c1b274ddc+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17602 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_2c9d7390c9520ec1c42e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17603 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_2c9d7390c9520ec1c42e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17604 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_2e0432da842e548ced4d+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17605 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_2e0432da842e548ced4d+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17606 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_30b02c94c40715ef1413+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17607 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_30b02c94c40715ef1413+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17608 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_33832659c4841cf2f06f+4a27301b/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17609 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_33832659c4841cf2f06f+4a27301b/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17610 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_33a2a1d2f2d8e8e49e1a+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17611 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_37b3ce1ee47af005bebe+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17612 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_38b27aaf11cf4045906c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17613 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_38b27aaf11cf4045906c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17614 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_38fae39fa3848a54e884+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17615 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_3a82adf89377a01449f1+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17616 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_3a82adf89377a01449f1+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17617 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_40c4614ec10cdf609375+1d5aa319/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17618 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_40c4614ec10cdf609375+1d5aa319/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17619 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_41a3183aaa9c2cbe3e28+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17620 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_41a3183aaa9c2cbe3e28+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17621 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4323097dc248ba27d8c6+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17622 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_442b55a691221bdb4b08+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17623 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_442b55a691221bdb4b08+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17624 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_44512bdd426602e62ce9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17625 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_44512bdd426602e62ce9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17626 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_45b435f72e0037ca46a2+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17627 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_497252034e9743a5ba18+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17628 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4a11e1ef3adf1ceb9513+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17629 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4a3055fc5c1fd109487a+1d5aa319/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17630 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4a3055fc5c1fd109487a+1d5aa319/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17631 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4b27b2b060ee68419d2a+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17632 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4b27b2b060ee68419d2a+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17633 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4c65baafff130ed529cb+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17634 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4df48492d35e4ffb8a16+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17635 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4e4db5fd3765679de86e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17636 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4e4db5fd3765679de86e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17637 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4f9febd8d1f222b8f798+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17638 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_4f9febd8d1f222b8f798+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17639 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_53f20d124198545c96d2+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17640 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_55d18f781d9212e2dfc6+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17641 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_57df875f7fa16817b32e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17642 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_57df875f7fa16817b32e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17643 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5a4dba7b006e51d8f5aa+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17644 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5df183a9a537186b793f/text_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17645 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5df183a9a537186b793f/text_encoder_2/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17646 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5df183a9a537186b793f/unet/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17647 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5df183a9a537186b793f/vae_decoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17648 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5df183a9a537186b793f/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17649 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5e0601a3382db348dc1e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17650 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_5e0601a3382db348dc1e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17651 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_601ad14244daafd4e761+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17652 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_617f751c1ee43436fa2c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17653 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_617f751c1ee43436fa2c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17654 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_61b53359a0a31e597238+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17655 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_63365ed7d18a04195fe5+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17656 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_672d3fac8fde8fb30afb+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17657 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_693e831be92621c50b6f+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17658 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6ac0045a775dbb747ff4+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17659 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6b694cb6a93ff1492ff7+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17660 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6b694cb6a93ff1492ff7+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17661 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6df920de2413af901e26+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17662 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6e73b50052e4c5001580+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17663 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6e73b50052e4c5001580+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17664 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6f5fa2b092920992ce0b+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17665 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_6f5fa2b092920992ce0b+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17666 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_706a16b21bfca50c4ce2+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17667 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_738b2e2b8defa3f16694+e457fe88/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17668 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73e765061ca0edc0984c/text_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17669 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73e765061ca0edc0984c/unet/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17670 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73e765061ca0edc0984c/vae_decoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17671 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73e765061ca0edc0984c/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17672 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73fde5a38839e503faa4+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17673 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_73fde5a38839e503faa4+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17674 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_797a0d607acbacd73bc4+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17675 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_7a143e9f44ff22eb2801+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17676 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_7a5ae276c555d2c5222b+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17677 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_7b01a92f1a36b7b0844c+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17678 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_7dcec5e93455378688b7+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17679 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_81ca1e738749cede64f5+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17680 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_84fac9181b5d14db36da+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17681 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_84fac9181b5d14db36da+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17682 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_850d83edca7646932466+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17683 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_850d83edca7646932466+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17684 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_885370837f47156d6058+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17685 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_885370837f47156d6058+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17686 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_8b9a3de21061e9ecce8a+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17687 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_8b9a3de21061e9ecce8a+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17688 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_93ed16e7c6730242b94b+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17689 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_93ed16e7c6730242b94b+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17690 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_942cb41a25c886b376c6+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17691 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_942cb41a25c886b376c6+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17692 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_943e620ba286661d8173+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17693 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_95b2a9888763feede71c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17694 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_95b2a9888763feede71c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17695 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_96b0595fcfb688e22991+1d5aa319/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17696 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_96b0595fcfb688e22991+1d5aa319/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17697 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_97e2a10e28dc6c0ad008+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17698 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_97ffaf0122e932464dcd+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17699 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_9e9ac2ab5c4f2d067d64+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17700 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_9f565ecac6bdf7c268d0+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17701 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_a1da31d69d33b0feed0d+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17702 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_a38d59aa68f454800914+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17703 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_a9abbb1d2ae349f44e9b+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17704 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ab49061cbff5ac26a535+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17705 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ab49061cbff5ac26a535+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17706 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_adb9128be89a1c98b633+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17707 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_addb5aa6d184bf31c794+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17708 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_addb5aa6d184bf31c794+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17709 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b209dc66669452006807+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17710 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b209dc66669452006807+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17711 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b352b056488c703f91b4+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17712 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b352b056488c703f91b4+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17713 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b4d25e616296d9d065d7+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17714 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b6ba419c618014159090+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17715 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b6ba419c618014159090+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17716 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b87d2eedb0cd1b974d71+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17717 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b87d2eedb0cd1b974d71+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17718 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b91ae604eae60b0fabf2+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17719 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_b91ae604eae60b0fabf2+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17720 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ba3b564583f21557a4a9+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17721 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_bd61d5d24b04539e47bb+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17722 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_bd61d5d24b04539e47bb+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17723 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_be22190e9ba1805e0bbd+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17724 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_bead9853fbf8905491da+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17725 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c0eb50bfe337a2ae0eb6+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17726 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c2339f7dfd92bc4f4a09+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17727 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c2339f7dfd92bc4f4a09+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17728 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c2719f909b7994638732+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17729 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c2719f909b7994638732+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17730 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c35c1d0c26d1633c765f+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17731 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c35c1d0c26d1633c765f+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17732 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c6310bffbfbf29e5d927/text_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17733 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c6310bffbfbf29e5d927/unet/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17734 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c6310bffbfbf29e5d927/vae_decoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17735 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c6310bffbfbf29e5d927/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
|
| 17736 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_c667425442470fb41cc2+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17737 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_cceaa2d0540a8dc32a5a+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17738 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_dccda32ed026f04a09f8+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17739 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_dd3c57cb3f607b586f3b+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17740 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_dd6d9cd7ecff6fd57d40+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17741 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_dd6d9cd7ecff6fd57d40+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17742 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_dedc330f2ecabe0f6b62+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17743 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e01d20c9f7f58e19f6fd+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17744 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e01d20c9f7f58e19f6fd+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17745 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e01d7b8d87f66fb4c5c4+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17746 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e01d7b8d87f66fb4c5c4+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17747 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e0e784ebf4b73fdc8a51+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17748 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e0e784ebf4b73fdc8a51+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17749 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e585b7ba3d40121c96d4+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17750 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e6f172aca340ebc38f62+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17751 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_e84ea5bc122d4d45ea24+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17752 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea1100f623c4d5b3dd18+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17753 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea1100f623c4d5b3dd18+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17754 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_eda9e5170145f93ff69b+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17755 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ee7d4038286fa40eb024+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17756 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ee7d4038286fa40eb024+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17757 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ef274caa64bc859f5c5c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17758 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_ef274caa64bc859f5c5c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17759 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_efd6f44c6d257085505e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17760 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_efd6f44c6d257085505e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17761 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f27c5b2a1477f1506e0c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17762 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f3069378727ea3b17875+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17763 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f3069378727ea3b17875+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17764 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f56602829db90683aac8+0788fa03/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17765 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f685396b88bb9384006c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17766 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f685396b88bb9384006c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17767 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f7c7d17145ae2991812c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17768 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_f7c7d17145ae2991812c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17769 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_fb55b8eb238996c22393+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17770 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_fb55b8eb238996c22393+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17771 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_fc09d37f2c192c8b245b+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 17772 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_fc09d37f2c192c8b245b+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 17773 |
+
neuronxcc-2.26.6360.0+6f180f47/MODULE_fe693526a669de0d2c0c+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/03b0c107d1cede36875199a5d51decfe04c473de2af9999f8577a028d74d0ab4/ec1560a24138c907085c.json
ADDED
|
@@ -0,0 +1,189 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolVLM-256M-Instruct",
|
| 4 |
+
"_task": "image-text-to-text",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Idefics3ForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"image_token_id": 49190,
|
| 10 |
+
"model_type": "idefics3",
|
| 11 |
+
"neuron": {
|
| 12 |
+
"_serialized_key": "NxDVLMNeuronConfig",
|
| 13 |
+
"batch_size": 2,
|
| 14 |
+
"capacity_factor": null,
|
| 15 |
+
"checkpoint_id": "HuggingFaceTB/SmolVLM-256M-Instruct",
|
| 16 |
+
"checkpoint_revision": "7e3e67edbbed1bf9888184d9df282b700a323964",
|
| 17 |
+
"continuous_batching": true,
|
| 18 |
+
"ep_degree": 1,
|
| 19 |
+
"fused_qkv": true,
|
| 20 |
+
"glu_mlp": true,
|
| 21 |
+
"image_seq_len": 64,
|
| 22 |
+
"image_size": 512,
|
| 23 |
+
"local_ranks_size": 2,
|
| 24 |
+
"max_batch_size": 2,
|
| 25 |
+
"max_context_length": 2048,
|
| 26 |
+
"max_num_images": 17,
|
| 27 |
+
"max_topk": 256,
|
| 28 |
+
"n_active_tokens": 2048,
|
| 29 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 30 |
+
"on_device_sampling": true,
|
| 31 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 32 |
+
"output_logits": false,
|
| 33 |
+
"pp_degree": 1,
|
| 34 |
+
"prefill_chunk_size": 1024,
|
| 35 |
+
"sequence_length": 2048,
|
| 36 |
+
"speculation_length": 0,
|
| 37 |
+
"start_rank_id": 0,
|
| 38 |
+
"target": "trn1",
|
| 39 |
+
"torch_dtype": "bfloat16",
|
| 40 |
+
"tp_degree": 2
|
| 41 |
+
},
|
| 42 |
+
"scale_factor": 4,
|
| 43 |
+
"text_config": {
|
| 44 |
+
"_attn_implementation_autoset": false,
|
| 45 |
+
"_flash_attn_2_enabled": true,
|
| 46 |
+
"_name_or_path": "None",
|
| 47 |
+
"architectures": [
|
| 48 |
+
"VLlama3ForCausalLM"
|
| 49 |
+
],
|
| 50 |
+
"attention_bias": false,
|
| 51 |
+
"attention_dropout": 0.0,
|
| 52 |
+
"dtype": "bfloat16",
|
| 53 |
+
"head_dim": 64,
|
| 54 |
+
"hidden_act": "silu",
|
| 55 |
+
"hidden_size": 576,
|
| 56 |
+
"initializer_range": 0.041666666666666664,
|
| 57 |
+
"intermediate_size": 1536,
|
| 58 |
+
"is_llama_config": true,
|
| 59 |
+
"max_position_embeddings": 8192,
|
| 60 |
+
"mlp_bias": false,
|
| 61 |
+
"model_type": "llama",
|
| 62 |
+
"neftune_noise_alpha": 0.0,
|
| 63 |
+
"num_attention_heads": 9,
|
| 64 |
+
"num_hidden_layers": 30,
|
| 65 |
+
"num_key_value_heads": 3,
|
| 66 |
+
"pad_token_id": 2,
|
| 67 |
+
"perceiver_config": {
|
| 68 |
+
"_attn_implementation_autoset": false,
|
| 69 |
+
"_name_or_path": "",
|
| 70 |
+
"add_cross_attention": false,
|
| 71 |
+
"architectures": null,
|
| 72 |
+
"attention_dropout": 0.0,
|
| 73 |
+
"bad_words_ids": null,
|
| 74 |
+
"begin_suppress_tokens": null,
|
| 75 |
+
"bos_token_id": null,
|
| 76 |
+
"chunk_size_feed_forward": 0,
|
| 77 |
+
"cross_attention_hidden_size": null,
|
| 78 |
+
"decoder_start_token_id": null,
|
| 79 |
+
"diversity_penalty": 0.0,
|
| 80 |
+
"do_sample": false,
|
| 81 |
+
"early_stopping": false,
|
| 82 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 83 |
+
"eos_token_id": null,
|
| 84 |
+
"exponential_decay_length_penalty": null,
|
| 85 |
+
"finetuning_task": null,
|
| 86 |
+
"forced_bos_token_id": null,
|
| 87 |
+
"forced_eos_token_id": null,
|
| 88 |
+
"hidden_act": "silu",
|
| 89 |
+
"id2label": {
|
| 90 |
+
"0": "LABEL_0",
|
| 91 |
+
"1": "LABEL_1"
|
| 92 |
+
},
|
| 93 |
+
"is_decoder": false,
|
| 94 |
+
"is_encoder_decoder": false,
|
| 95 |
+
"label2id": {
|
| 96 |
+
"LABEL_0": 0,
|
| 97 |
+
"LABEL_1": 1
|
| 98 |
+
},
|
| 99 |
+
"length_penalty": 1.0,
|
| 100 |
+
"max_length": 20,
|
| 101 |
+
"min_length": 0,
|
| 102 |
+
"model_type": "vllama3",
|
| 103 |
+
"no_repeat_ngram_size": 0,
|
| 104 |
+
"num_beam_groups": 1,
|
| 105 |
+
"num_beams": 1,
|
| 106 |
+
"num_key_value_heads": 1,
|
| 107 |
+
"num_return_sequences": 1,
|
| 108 |
+
"output_attentions": false,
|
| 109 |
+
"output_hidden_states": false,
|
| 110 |
+
"output_scores": false,
|
| 111 |
+
"pad_token_id": null,
|
| 112 |
+
"prefix": null,
|
| 113 |
+
"problem_type": null,
|
| 114 |
+
"pruned_heads": {},
|
| 115 |
+
"qk_layer_norms_perceiver": false,
|
| 116 |
+
"remove_invalid_values": false,
|
| 117 |
+
"repetition_penalty": 1.0,
|
| 118 |
+
"resampler_depth": 6,
|
| 119 |
+
"resampler_head_dim": 96,
|
| 120 |
+
"resampler_n_heads": 16,
|
| 121 |
+
"resampler_n_latents": 64,
|
| 122 |
+
"return_dict": true,
|
| 123 |
+
"return_dict_in_generate": false,
|
| 124 |
+
"sep_token_id": null,
|
| 125 |
+
"suppress_tokens": null,
|
| 126 |
+
"task_specific_params": null,
|
| 127 |
+
"temperature": 1.0,
|
| 128 |
+
"tf_legacy_loss": false,
|
| 129 |
+
"tie_encoder_decoder": false,
|
| 130 |
+
"tie_word_embeddings": true,
|
| 131 |
+
"tokenizer_class": null,
|
| 132 |
+
"top_k": 50,
|
| 133 |
+
"top_p": 1.0,
|
| 134 |
+
"torch_dtype": null,
|
| 135 |
+
"torchscript": false,
|
| 136 |
+
"transformers_version": "4.46.0",
|
| 137 |
+
"typical_p": 1.0,
|
| 138 |
+
"use_bfloat16": false
|
| 139 |
+
},
|
| 140 |
+
"pixel_shuffle_factor": 4,
|
| 141 |
+
"pretraining_tp": 1,
|
| 142 |
+
"qk_layer_norms": false,
|
| 143 |
+
"rms_norm_eps": 1e-05,
|
| 144 |
+
"rope_interleaved": false,
|
| 145 |
+
"rope_scaling": null,
|
| 146 |
+
"rope_theta": 100000,
|
| 147 |
+
"transformers.js_config": {
|
| 148 |
+
"kv_cache_dtype": {
|
| 149 |
+
"fp16": "float16",
|
| 150 |
+
"q4f16": "float16"
|
| 151 |
+
}
|
| 152 |
+
},
|
| 153 |
+
"use_cache": true,
|
| 154 |
+
"use_resampler": false,
|
| 155 |
+
"vocab_size": 49280
|
| 156 |
+
},
|
| 157 |
+
"tie_word_embeddings": false,
|
| 158 |
+
"transformers.js_config": {
|
| 159 |
+
"kv_cache_dtype": {
|
| 160 |
+
"fp16": "float16",
|
| 161 |
+
"q4f16": "float16"
|
| 162 |
+
}
|
| 163 |
+
},
|
| 164 |
+
"use_cache": true,
|
| 165 |
+
"vision_config": {
|
| 166 |
+
"_attn_implementation_autoset": false,
|
| 167 |
+
"attention_dropout": 0.0,
|
| 168 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 169 |
+
"hidden_size": 768,
|
| 170 |
+
"image_size": 512,
|
| 171 |
+
"initializer_range": 0.02,
|
| 172 |
+
"intermediate_size": 3072,
|
| 173 |
+
"layer_norm_eps": 1e-06,
|
| 174 |
+
"max_image_size": {
|
| 175 |
+
"longest_edge": 512
|
| 176 |
+
},
|
| 177 |
+
"model_type": "idefics3_vision",
|
| 178 |
+
"num_attention_heads": 12,
|
| 179 |
+
"num_channels": 3,
|
| 180 |
+
"num_hidden_layers": 12,
|
| 181 |
+
"patch_size": 16,
|
| 182 |
+
"size": {
|
| 183 |
+
"longest_edge": 2048
|
| 184 |
+
},
|
| 185 |
+
"tie_word_embeddings": false,
|
| 186 |
+
"use_base_siglip": true
|
| 187 |
+
},
|
| 188 |
+
"vocab_size": 49280
|
| 189 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/04e22cfac37c408caa51d61b493ee185fc351c2f9db86171e024b5d0ad5b25c1/c56a8bce0577ce31a594.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-qwen3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention"
|
| 19 |
+
],
|
| 20 |
+
"max_position_embeddings": 32768,
|
| 21 |
+
"max_window_layers": 36,
|
| 22 |
+
"model_type": "qwen3",
|
| 23 |
+
"neuron": {
|
| 24 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 25 |
+
"batch_size": 2,
|
| 26 |
+
"capacity_factor": null,
|
| 27 |
+
"checkpoint_id": "llamafactory/tiny-random-qwen3",
|
| 28 |
+
"checkpoint_revision": "81d6f5f5e05ed53ea8a1d19431266a486e46bbd8",
|
| 29 |
+
"continuous_batching": true,
|
| 30 |
+
"ep_degree": 1,
|
| 31 |
+
"fused_qkv": true,
|
| 32 |
+
"glu_mlp": true,
|
| 33 |
+
"local_ranks_size": 2,
|
| 34 |
+
"max_batch_size": 2,
|
| 35 |
+
"max_context_length": 1024,
|
| 36 |
+
"max_topk": 256,
|
| 37 |
+
"n_active_tokens": 1024,
|
| 38 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 39 |
+
"on_device_sampling": false,
|
| 40 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 41 |
+
"output_logits": false,
|
| 42 |
+
"pp_degree": 1,
|
| 43 |
+
"prefill_chunk_size": 0,
|
| 44 |
+
"sequence_length": 1024,
|
| 45 |
+
"speculation_length": 0,
|
| 46 |
+
"start_rank_id": 0,
|
| 47 |
+
"target": "trn1",
|
| 48 |
+
"torch_dtype": "bfloat16",
|
| 49 |
+
"tp_degree": 2
|
| 50 |
+
},
|
| 51 |
+
"num_attention_heads": 4,
|
| 52 |
+
"num_hidden_layers": 2,
|
| 53 |
+
"num_key_value_heads": 4,
|
| 54 |
+
"rms_norm_eps": 1e-06,
|
| 55 |
+
"rope_scaling": null,
|
| 56 |
+
"rope_theta": 5000000,
|
| 57 |
+
"sliding_window": null,
|
| 58 |
+
"tie_word_embeddings": true,
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"use_sliding_window": false,
|
| 61 |
+
"vocab_size": 151936
|
| 62 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/0cc25526e2cfc37a8875a3752f33c4d7505d8a07b869d0f3f41915cf6e763b74/2132439b615164bef4eb.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 12 |
+
"embedding_multiplier": 12.0,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 2048,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 8192,
|
| 17 |
+
"logits_scaling": 8.0,
|
| 18 |
+
"max_position_embeddings": 131072,
|
| 19 |
+
"mlp_bias": false,
|
| 20 |
+
"model_type": "granite",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 23 |
+
"batch_size": 4,
|
| 24 |
+
"capacity_factor": null,
|
| 25 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
| 26 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
| 27 |
+
"continuous_batching": true,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": true,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 2,
|
| 32 |
+
"max_batch_size": 4,
|
| 33 |
+
"max_context_length": 1024,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 1024,
|
| 36 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"prefill_chunk_size": 0,
|
| 42 |
+
"sequence_length": 1024,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": "trn1",
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"tp_degree": 2
|
| 48 |
+
},
|
| 49 |
+
"num_attention_heads": 32,
|
| 50 |
+
"num_hidden_layers": 40,
|
| 51 |
+
"num_key_value_heads": 8,
|
| 52 |
+
"residual_multiplier": 0.22,
|
| 53 |
+
"rms_norm_eps": 1e-05,
|
| 54 |
+
"rope_scaling": null,
|
| 55 |
+
"rope_theta": 5000000.0,
|
| 56 |
+
"tie_word_embeddings": true,
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 49155
|
| 59 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/0cc25526e2cfc37a8875a3752f33c4d7505d8a07b869d0f3f41915cf6e763b74/da74b3645f29f4218169.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 12 |
+
"embedding_multiplier": 12.0,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 2048,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 8192,
|
| 17 |
+
"logits_scaling": 8.0,
|
| 18 |
+
"max_position_embeddings": 131072,
|
| 19 |
+
"mlp_bias": false,
|
| 20 |
+
"model_type": "granite",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 23 |
+
"batch_size": 1,
|
| 24 |
+
"capacity_factor": null,
|
| 25 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
| 26 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
| 27 |
+
"continuous_batching": false,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": true,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 2,
|
| 32 |
+
"max_batch_size": 1,
|
| 33 |
+
"max_context_length": 8192,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 8192,
|
| 36 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"prefill_chunk_size": 1024,
|
| 42 |
+
"sequence_length": 8192,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": "trn1",
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"tp_degree": 2
|
| 48 |
+
},
|
| 49 |
+
"num_attention_heads": 32,
|
| 50 |
+
"num_hidden_layers": 40,
|
| 51 |
+
"num_key_value_heads": 8,
|
| 52 |
+
"residual_multiplier": 0.22,
|
| 53 |
+
"rms_norm_eps": 1e-05,
|
| 54 |
+
"rope_scaling": null,
|
| 55 |
+
"rope_theta": 5000000.0,
|
| 56 |
+
"tie_word_embeddings": true,
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 49155
|
| 59 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/6ed6aa77ee480d24287c.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-270m-it",
|
| 4 |
+
"_sliding_window_pattern": 6,
|
| 5 |
+
"_task": "text-generation",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"Gemma3ForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"attn_logit_softcapping": null,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"final_logit_softcapping": null,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 640,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 2048,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"sliding_attention",
|
| 21 |
+
"sliding_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"sliding_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"sliding_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"sliding_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"full_attention"
|
| 38 |
+
],
|
| 39 |
+
"max_position_embeddings": 32768,
|
| 40 |
+
"model_type": "gemma3_text",
|
| 41 |
+
"neuron": {
|
| 42 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 43 |
+
"batch_size": 1,
|
| 44 |
+
"capacity_factor": null,
|
| 45 |
+
"checkpoint_id": "unsloth/gemma-3-270m-it",
|
| 46 |
+
"checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
|
| 47 |
+
"continuous_batching": false,
|
| 48 |
+
"ep_degree": 1,
|
| 49 |
+
"fused_qkv": true,
|
| 50 |
+
"glu_mlp": true,
|
| 51 |
+
"local_ranks_size": 2,
|
| 52 |
+
"max_batch_size": 1,
|
| 53 |
+
"max_context_length": 8192,
|
| 54 |
+
"max_topk": 256,
|
| 55 |
+
"n_active_tokens": 8192,
|
| 56 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 57 |
+
"on_device_sampling": true,
|
| 58 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 59 |
+
"output_logits": true,
|
| 60 |
+
"pp_degree": 1,
|
| 61 |
+
"prefill_chunk_size": 0,
|
| 62 |
+
"sequence_length": 8192,
|
| 63 |
+
"speculation_length": 0,
|
| 64 |
+
"start_rank_id": 0,
|
| 65 |
+
"target": "trn1",
|
| 66 |
+
"torch_dtype": "bfloat16",
|
| 67 |
+
"tp_degree": 2
|
| 68 |
+
},
|
| 69 |
+
"num_attention_heads": 4,
|
| 70 |
+
"num_hidden_layers": 18,
|
| 71 |
+
"num_key_value_heads": 1,
|
| 72 |
+
"query_pre_attn_scalar": 256,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_local_base_freq": 10000.0,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": 512,
|
| 78 |
+
"unsloth_fixed": true,
|
| 79 |
+
"use_bidirectional_attention": false,
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 262144
|
| 82 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/748542be1bcc5fe3cdf1.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-270m-it",
|
| 4 |
+
"_sliding_window_pattern": 6,
|
| 5 |
+
"_task": "text-generation",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"Gemma3ForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"attn_logit_softcapping": null,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"final_logit_softcapping": null,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 640,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 2048,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"sliding_attention",
|
| 21 |
+
"sliding_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"sliding_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"sliding_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"sliding_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"full_attention"
|
| 38 |
+
],
|
| 39 |
+
"max_position_embeddings": 32768,
|
| 40 |
+
"model_type": "gemma3_text",
|
| 41 |
+
"neuron": {
|
| 42 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 43 |
+
"batch_size": 1,
|
| 44 |
+
"capacity_factor": null,
|
| 45 |
+
"checkpoint_id": "unsloth/gemma-3-270m-it",
|
| 46 |
+
"checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
|
| 47 |
+
"continuous_batching": false,
|
| 48 |
+
"ep_degree": 1,
|
| 49 |
+
"fused_qkv": true,
|
| 50 |
+
"glu_mlp": true,
|
| 51 |
+
"local_ranks_size": 2,
|
| 52 |
+
"max_batch_size": 1,
|
| 53 |
+
"max_context_length": 8192,
|
| 54 |
+
"max_topk": 256,
|
| 55 |
+
"n_active_tokens": 8192,
|
| 56 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 57 |
+
"on_device_sampling": true,
|
| 58 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 59 |
+
"output_logits": false,
|
| 60 |
+
"pp_degree": 1,
|
| 61 |
+
"prefill_chunk_size": 0,
|
| 62 |
+
"sequence_length": 8192,
|
| 63 |
+
"speculation_length": 0,
|
| 64 |
+
"start_rank_id": 0,
|
| 65 |
+
"target": "trn1",
|
| 66 |
+
"torch_dtype": "bfloat16",
|
| 67 |
+
"tp_degree": 2
|
| 68 |
+
},
|
| 69 |
+
"num_attention_heads": 4,
|
| 70 |
+
"num_hidden_layers": 18,
|
| 71 |
+
"num_key_value_heads": 1,
|
| 72 |
+
"query_pre_attn_scalar": 256,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_local_base_freq": 10000.0,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": 512,
|
| 78 |
+
"unsloth_fixed": true,
|
| 79 |
+
"use_bidirectional_attention": false,
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 262144
|
| 82 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/9a130a0e2b7177d1dea1.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-270m-it",
|
| 4 |
+
"_sliding_window_pattern": 6,
|
| 5 |
+
"_task": "text-generation",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"Gemma3ForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"attn_logit_softcapping": null,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"final_logit_softcapping": null,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 640,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 2048,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"sliding_attention",
|
| 21 |
+
"sliding_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"sliding_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"sliding_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"sliding_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"full_attention"
|
| 38 |
+
],
|
| 39 |
+
"max_position_embeddings": 32768,
|
| 40 |
+
"model_type": "gemma3_text",
|
| 41 |
+
"neuron": {
|
| 42 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 43 |
+
"batch_size": 4,
|
| 44 |
+
"capacity_factor": null,
|
| 45 |
+
"checkpoint_id": "unsloth/gemma-3-270m-it",
|
| 46 |
+
"checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
|
| 47 |
+
"continuous_batching": true,
|
| 48 |
+
"ep_degree": 1,
|
| 49 |
+
"fused_qkv": true,
|
| 50 |
+
"glu_mlp": true,
|
| 51 |
+
"local_ranks_size": 2,
|
| 52 |
+
"max_batch_size": 4,
|
| 53 |
+
"max_context_length": 1024,
|
| 54 |
+
"max_topk": 256,
|
| 55 |
+
"n_active_tokens": 1024,
|
| 56 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 57 |
+
"on_device_sampling": true,
|
| 58 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 59 |
+
"output_logits": false,
|
| 60 |
+
"pp_degree": 1,
|
| 61 |
+
"prefill_chunk_size": 0,
|
| 62 |
+
"sequence_length": 1024,
|
| 63 |
+
"speculation_length": 0,
|
| 64 |
+
"start_rank_id": 0,
|
| 65 |
+
"target": "trn1",
|
| 66 |
+
"torch_dtype": "bfloat16",
|
| 67 |
+
"tp_degree": 2
|
| 68 |
+
},
|
| 69 |
+
"num_attention_heads": 4,
|
| 70 |
+
"num_hidden_layers": 18,
|
| 71 |
+
"num_key_value_heads": 1,
|
| 72 |
+
"query_pre_attn_scalar": 256,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_local_base_freq": 10000.0,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": 512,
|
| 78 |
+
"unsloth_fixed": true,
|
| 79 |
+
"use_bidirectional_attention": false,
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 262144
|
| 82 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/196ef14bd9a3e0d3035c.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 896,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 4864,
|
| 14 |
+
"layer_types": [
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention"
|
| 39 |
+
],
|
| 40 |
+
"max_position_embeddings": 32768,
|
| 41 |
+
"max_window_layers": 24,
|
| 42 |
+
"model_type": "qwen2",
|
| 43 |
+
"neuron": {
|
| 44 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 45 |
+
"batch_size": 1,
|
| 46 |
+
"capacity_factor": null,
|
| 47 |
+
"checkpoint_id": "Qwen/Qwen2.5-0.5B",
|
| 48 |
+
"checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
|
| 49 |
+
"continuous_batching": false,
|
| 50 |
+
"ep_degree": 1,
|
| 51 |
+
"fused_qkv": false,
|
| 52 |
+
"glu_mlp": true,
|
| 53 |
+
"local_ranks_size": 2,
|
| 54 |
+
"max_batch_size": 1,
|
| 55 |
+
"max_context_length": 8192,
|
| 56 |
+
"max_topk": 256,
|
| 57 |
+
"n_active_tokens": 8192,
|
| 58 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 59 |
+
"on_device_sampling": true,
|
| 60 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 61 |
+
"output_logits": false,
|
| 62 |
+
"pp_degree": 1,
|
| 63 |
+
"prefill_chunk_size": 1024,
|
| 64 |
+
"sequence_length": 8192,
|
| 65 |
+
"speculation_length": 0,
|
| 66 |
+
"start_rank_id": 0,
|
| 67 |
+
"target": "trn1",
|
| 68 |
+
"torch_dtype": "bfloat16",
|
| 69 |
+
"tp_degree": 2
|
| 70 |
+
},
|
| 71 |
+
"num_attention_heads": 14,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"num_key_value_heads": 2,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": null,
|
| 78 |
+
"tie_word_embeddings": true,
|
| 79 |
+
"use_cache": true,
|
| 80 |
+
"use_mrope": false,
|
| 81 |
+
"use_sliding_window": false,
|
| 82 |
+
"vocab_size": 151936
|
| 83 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/3474ddeda6545f7e5db7.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 896,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 4864,
|
| 14 |
+
"layer_types": [
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention"
|
| 39 |
+
],
|
| 40 |
+
"max_position_embeddings": 32768,
|
| 41 |
+
"max_window_layers": 24,
|
| 42 |
+
"model_type": "qwen2",
|
| 43 |
+
"neuron": {
|
| 44 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 45 |
+
"batch_size": 4,
|
| 46 |
+
"capacity_factor": null,
|
| 47 |
+
"checkpoint_id": "Qwen/Qwen2.5-0.5B",
|
| 48 |
+
"checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
|
| 49 |
+
"continuous_batching": true,
|
| 50 |
+
"ep_degree": 1,
|
| 51 |
+
"fused_qkv": false,
|
| 52 |
+
"glu_mlp": true,
|
| 53 |
+
"local_ranks_size": 2,
|
| 54 |
+
"max_batch_size": 4,
|
| 55 |
+
"max_context_length": 1024,
|
| 56 |
+
"max_topk": 256,
|
| 57 |
+
"n_active_tokens": 1024,
|
| 58 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 59 |
+
"on_device_sampling": false,
|
| 60 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 61 |
+
"output_logits": false,
|
| 62 |
+
"pp_degree": 1,
|
| 63 |
+
"prefill_chunk_size": 0,
|
| 64 |
+
"sequence_length": 1024,
|
| 65 |
+
"speculation_length": 0,
|
| 66 |
+
"start_rank_id": 0,
|
| 67 |
+
"target": "trn1",
|
| 68 |
+
"torch_dtype": "bfloat16",
|
| 69 |
+
"tp_degree": 2
|
| 70 |
+
},
|
| 71 |
+
"num_attention_heads": 14,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"num_key_value_heads": 2,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": null,
|
| 78 |
+
"tie_word_embeddings": true,
|
| 79 |
+
"use_cache": true,
|
| 80 |
+
"use_mrope": false,
|
| 81 |
+
"use_sliding_window": false,
|
| 82 |
+
"vocab_size": 151936
|
| 83 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/0d1cbffb3e66b482c261.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 32768,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 6,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "97b0c614be4d77ee51c0cef4e5f07c00f9eb65b3",
|
| 55 |
+
"continuous_batching": false,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 6,
|
| 61 |
+
"max_context_length": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 1024,
|
| 70 |
+
"sequence_length": 8192,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151669
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/61ff8eae2078e66cabfd.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 32768,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 4,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "97b0c614be4d77ee51c0cef4e5f07c00f9eb65b3",
|
| 55 |
+
"continuous_batching": false,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 1024,
|
| 70 |
+
"sequence_length": 8192,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151669
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/24892d985c7b4406e3d9.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-Llama-3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
| 24 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 1024,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "float16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 4,
|
| 48 |
+
"num_hidden_layers": 2,
|
| 49 |
+
"num_key_value_heads": 4,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 8.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": false,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/dedf7a5ae53bb6e1b3fc.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-Llama-3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 2,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
| 24 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
| 25 |
+
"continuous_batching": true,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 2,
|
| 31 |
+
"max_context_length": 128,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 128,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 128,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn2",
|
| 44 |
+
"torch_dtype": "float16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 4,
|
| 48 |
+
"num_hidden_layers": 2,
|
| 49 |
+
"num_key_value_heads": 4,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 8.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": false,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/6454afdf3e9d66c7226c13a575b718845c25e53b0699600ba2bb4f883e9d841b/df8b960809b8e9d97bd2.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-Llama-3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
| 24 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 512,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 512,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 512,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "float16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 4,
|
| 48 |
+
"num_hidden_layers": 2,
|
| 49 |
+
"num_key_value_heads": 4,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 8.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": false,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/73707b485eab9008c7aba7f5dad0ce2384ac685318d5f888c12fa0d81ed90b19/309698d1c71c3e0c8eb8.json
ADDED
|
@@ -0,0 +1,135 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolLM3-3B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"SmolLM3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 65536,
|
| 54 |
+
"max_window_layers": 28,
|
| 55 |
+
"mlp_bias": false,
|
| 56 |
+
"model_type": "smollm3",
|
| 57 |
+
"neuron": {
|
| 58 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 59 |
+
"batch_size": 1,
|
| 60 |
+
"capacity_factor": null,
|
| 61 |
+
"checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
|
| 62 |
+
"checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
|
| 63 |
+
"continuous_batching": false,
|
| 64 |
+
"ep_degree": 1,
|
| 65 |
+
"fused_qkv": true,
|
| 66 |
+
"glu_mlp": true,
|
| 67 |
+
"local_ranks_size": 2,
|
| 68 |
+
"max_batch_size": 1,
|
| 69 |
+
"max_context_length": 8192,
|
| 70 |
+
"max_topk": 256,
|
| 71 |
+
"n_active_tokens": 8192,
|
| 72 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 73 |
+
"on_device_sampling": true,
|
| 74 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 75 |
+
"output_logits": false,
|
| 76 |
+
"pp_degree": 1,
|
| 77 |
+
"prefill_chunk_size": 1024,
|
| 78 |
+
"sequence_length": 8192,
|
| 79 |
+
"speculation_length": 0,
|
| 80 |
+
"start_rank_id": 0,
|
| 81 |
+
"target": "trn1",
|
| 82 |
+
"torch_dtype": "bfloat16",
|
| 83 |
+
"tp_degree": 2
|
| 84 |
+
},
|
| 85 |
+
"no_rope_layer_interval": 4,
|
| 86 |
+
"no_rope_layers": [
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
1,
|
| 90 |
+
0,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
1,
|
| 94 |
+
0,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
1,
|
| 98 |
+
0,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
1,
|
| 102 |
+
0,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
1,
|
| 106 |
+
0,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
1,
|
| 110 |
+
0,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
1,
|
| 114 |
+
0,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
1,
|
| 118 |
+
0,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
1,
|
| 122 |
+
0
|
| 123 |
+
],
|
| 124 |
+
"num_attention_heads": 16,
|
| 125 |
+
"num_hidden_layers": 36,
|
| 126 |
+
"num_key_value_heads": 4,
|
| 127 |
+
"pretraining_tp": 2,
|
| 128 |
+
"rms_norm_eps": 1e-06,
|
| 129 |
+
"rope_scaling": null,
|
| 130 |
+
"rope_theta": 5000000.0,
|
| 131 |
+
"sliding_window": null,
|
| 132 |
+
"use_cache": false,
|
| 133 |
+
"use_sliding_window": false,
|
| 134 |
+
"vocab_size": 128256
|
| 135 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/73707b485eab9008c7aba7f5dad0ce2384ac685318d5f888c12fa0d81ed90b19/8d67a5db932b99602641.json
ADDED
|
@@ -0,0 +1,135 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolLM3-3B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"SmolLM3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 65536,
|
| 54 |
+
"max_window_layers": 28,
|
| 55 |
+
"mlp_bias": false,
|
| 56 |
+
"model_type": "smollm3",
|
| 57 |
+
"neuron": {
|
| 58 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 59 |
+
"batch_size": 4,
|
| 60 |
+
"capacity_factor": null,
|
| 61 |
+
"checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
|
| 62 |
+
"checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
|
| 63 |
+
"continuous_batching": true,
|
| 64 |
+
"ep_degree": 1,
|
| 65 |
+
"fused_qkv": true,
|
| 66 |
+
"glu_mlp": true,
|
| 67 |
+
"local_ranks_size": 2,
|
| 68 |
+
"max_batch_size": 4,
|
| 69 |
+
"max_context_length": 1024,
|
| 70 |
+
"max_topk": 256,
|
| 71 |
+
"n_active_tokens": 1024,
|
| 72 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 73 |
+
"on_device_sampling": true,
|
| 74 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 75 |
+
"output_logits": false,
|
| 76 |
+
"pp_degree": 1,
|
| 77 |
+
"prefill_chunk_size": 0,
|
| 78 |
+
"sequence_length": 1024,
|
| 79 |
+
"speculation_length": 0,
|
| 80 |
+
"start_rank_id": 0,
|
| 81 |
+
"target": "trn1",
|
| 82 |
+
"torch_dtype": "bfloat16",
|
| 83 |
+
"tp_degree": 2
|
| 84 |
+
},
|
| 85 |
+
"no_rope_layer_interval": 4,
|
| 86 |
+
"no_rope_layers": [
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
1,
|
| 90 |
+
0,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
1,
|
| 94 |
+
0,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
1,
|
| 98 |
+
0,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
1,
|
| 102 |
+
0,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
1,
|
| 106 |
+
0,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
1,
|
| 110 |
+
0,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
1,
|
| 114 |
+
0,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
1,
|
| 118 |
+
0,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
1,
|
| 122 |
+
0
|
| 123 |
+
],
|
| 124 |
+
"num_attention_heads": 16,
|
| 125 |
+
"num_hidden_layers": 36,
|
| 126 |
+
"num_key_value_heads": 4,
|
| 127 |
+
"pretraining_tp": 2,
|
| 128 |
+
"rms_norm_eps": 1e-06,
|
| 129 |
+
"rope_scaling": null,
|
| 130 |
+
"rope_theta": 5000000.0,
|
| 131 |
+
"sliding_window": null,
|
| 132 |
+
"use_cache": false,
|
| 133 |
+
"use_sliding_window": false,
|
| 134 |
+
"vocab_size": 128256
|
| 135 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/0b70cb2cd9d99c1bdf34.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 40960,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 4,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": true,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 1024,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 1024,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 0,
|
| 70 |
+
"sequence_length": 1024,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/805a8935ea4fbf9c4df6.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 40960,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 4,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": true,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 1,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 1024,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 1024,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": true,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 0,
|
| 70 |
+
"sequence_length": 1024,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 1
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/fd610699b7c75d5fa5cc.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 40960,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 1,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": false,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 1,
|
| 61 |
+
"max_context_length": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": true,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 1024,
|
| 70 |
+
"sequence_length": 8192,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/5df183a9a537186b793f.json
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "MultiModelCacheEntry",
|
| 3 |
+
"_model_id": "/home/tengomucho/.cache/huggingface/hub/models--echarlaix--tiny-random-stable-diffusion-xl/snapshots/434a0025554940d46ddd5dd10f0abc2788076b47/vae",
|
| 4 |
+
"_task": null,
|
| 5 |
+
"text_encoder": {
|
| 6 |
+
"architectures": [
|
| 7 |
+
"CLIPTextModel"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"export_model_type": "clip-text-model",
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_size": 32,
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 37,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 77,
|
| 19 |
+
"model_type": "clip_text_model",
|
| 20 |
+
"neuron": {
|
| 21 |
+
"auto_cast": "matmul",
|
| 22 |
+
"auto_cast_type": "bf16",
|
| 23 |
+
"compiler_type": "neuronx-cc",
|
| 24 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 25 |
+
"dynamic_batch_size": false,
|
| 26 |
+
"float_dtype": "fp32",
|
| 27 |
+
"inline_weights_to_neff": true,
|
| 28 |
+
"instance_type": "inf2",
|
| 29 |
+
"int_dtype": "int64",
|
| 30 |
+
"optlevel": "2",
|
| 31 |
+
"output_attentions": false,
|
| 32 |
+
"output_hidden_states": false,
|
| 33 |
+
"static_batch_size": 1,
|
| 34 |
+
"static_sequence_length": 77,
|
| 35 |
+
"task": "feature-extraction",
|
| 36 |
+
"tensor_parallel_size": 1
|
| 37 |
+
},
|
| 38 |
+
"num_attention_heads": 4,
|
| 39 |
+
"num_hidden_layers": 5,
|
| 40 |
+
"output_hidden_states": true,
|
| 41 |
+
"vocab_size": 1000
|
| 42 |
+
},
|
| 43 |
+
"text_encoder_2": {
|
| 44 |
+
"architectures": [
|
| 45 |
+
"CLIPTextModelWithProjection"
|
| 46 |
+
],
|
| 47 |
+
"attention_dropout": 0.0,
|
| 48 |
+
"dtype": "float32",
|
| 49 |
+
"export_model_type": "clip-text-with-projection",
|
| 50 |
+
"hidden_act": "gelu",
|
| 51 |
+
"hidden_size": 32,
|
| 52 |
+
"initializer_factor": 1.0,
|
| 53 |
+
"initializer_range": 0.02,
|
| 54 |
+
"intermediate_size": 37,
|
| 55 |
+
"layer_norm_eps": 1e-05,
|
| 56 |
+
"max_position_embeddings": 77,
|
| 57 |
+
"model_type": "clip_text_model",
|
| 58 |
+
"neuron": {
|
| 59 |
+
"auto_cast": "matmul",
|
| 60 |
+
"auto_cast_type": "bf16",
|
| 61 |
+
"compiler_type": "neuronx-cc",
|
| 62 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 63 |
+
"dynamic_batch_size": false,
|
| 64 |
+
"float_dtype": "fp32",
|
| 65 |
+
"inline_weights_to_neff": true,
|
| 66 |
+
"instance_type": "inf2",
|
| 67 |
+
"int_dtype": "int64",
|
| 68 |
+
"optlevel": "2",
|
| 69 |
+
"output_attentions": false,
|
| 70 |
+
"output_hidden_states": false,
|
| 71 |
+
"static_batch_size": 1,
|
| 72 |
+
"static_sequence_length": 77,
|
| 73 |
+
"task": "feature-extraction",
|
| 74 |
+
"tensor_parallel_size": 1
|
| 75 |
+
},
|
| 76 |
+
"num_attention_heads": 4,
|
| 77 |
+
"num_hidden_layers": 5,
|
| 78 |
+
"output_hidden_states": true,
|
| 79 |
+
"vocab_size": 1000
|
| 80 |
+
},
|
| 81 |
+
"unet": {
|
| 82 |
+
"_class_name": "UNet2DConditionModel",
|
| 83 |
+
"act_fn": "silu",
|
| 84 |
+
"addition_embed_type": "text_time",
|
| 85 |
+
"addition_embed_type_num_heads": 64,
|
| 86 |
+
"addition_time_embed_dim": 8,
|
| 87 |
+
"attention_head_dim": [
|
| 88 |
+
2,
|
| 89 |
+
4
|
| 90 |
+
],
|
| 91 |
+
"attention_type": "default",
|
| 92 |
+
"block_out_channels": [
|
| 93 |
+
32,
|
| 94 |
+
64
|
| 95 |
+
],
|
| 96 |
+
"center_input_sample": false,
|
| 97 |
+
"class_embed_type": null,
|
| 98 |
+
"class_embeddings_concat": false,
|
| 99 |
+
"conv_in_kernel": 3,
|
| 100 |
+
"conv_out_kernel": 3,
|
| 101 |
+
"cross_attention_dim": 64,
|
| 102 |
+
"cross_attention_norm": null,
|
| 103 |
+
"down_block_types": [
|
| 104 |
+
"DownBlock2D",
|
| 105 |
+
"CrossAttnDownBlock2D"
|
| 106 |
+
],
|
| 107 |
+
"downsample_padding": 1,
|
| 108 |
+
"dropout": 0.0,
|
| 109 |
+
"dual_cross_attention": false,
|
| 110 |
+
"encoder_hid_dim": null,
|
| 111 |
+
"encoder_hid_dim_type": null,
|
| 112 |
+
"flip_sin_to_cos": true,
|
| 113 |
+
"freq_shift": 0,
|
| 114 |
+
"in_channels": 4,
|
| 115 |
+
"layers_per_block": 2,
|
| 116 |
+
"mid_block_only_cross_attention": null,
|
| 117 |
+
"mid_block_scale_factor": 1,
|
| 118 |
+
"mid_block_type": "UNetMidBlock2DCrossAttn",
|
| 119 |
+
"neuron": {
|
| 120 |
+
"auto_cast": "matmul",
|
| 121 |
+
"auto_cast_type": "bf16",
|
| 122 |
+
"compiler_type": "neuronx-cc",
|
| 123 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 124 |
+
"dynamic_batch_size": false,
|
| 125 |
+
"float_dtype": "fp32",
|
| 126 |
+
"inline_weights_to_neff": true,
|
| 127 |
+
"instance_type": "inf2",
|
| 128 |
+
"int_dtype": "int64",
|
| 129 |
+
"optlevel": "2",
|
| 130 |
+
"output_attentions": false,
|
| 131 |
+
"output_hidden_states": false,
|
| 132 |
+
"static_batch_size": 1,
|
| 133 |
+
"static_height": 32,
|
| 134 |
+
"static_num_channels": 4,
|
| 135 |
+
"static_sequence_length": 77,
|
| 136 |
+
"static_vae_scale_factor": 2,
|
| 137 |
+
"static_width": 32,
|
| 138 |
+
"task": "semantic-segmentation",
|
| 139 |
+
"tensor_parallel_size": 1
|
| 140 |
+
},
|
| 141 |
+
"norm_eps": 1e-05,
|
| 142 |
+
"norm_num_groups": 32,
|
| 143 |
+
"num_attention_heads": null,
|
| 144 |
+
"num_class_embeds": null,
|
| 145 |
+
"only_cross_attention": false,
|
| 146 |
+
"out_channels": 4,
|
| 147 |
+
"output_attentions": false,
|
| 148 |
+
"projection_class_embeddings_input_dim": 80,
|
| 149 |
+
"resnet_out_scale_factor": 1.0,
|
| 150 |
+
"resnet_skip_time_act": false,
|
| 151 |
+
"resnet_time_scale_shift": "default",
|
| 152 |
+
"reverse_transformer_layers_per_block": null,
|
| 153 |
+
"time_cond_proj_dim": null,
|
| 154 |
+
"time_embedding_act_fn": null,
|
| 155 |
+
"time_embedding_dim": null,
|
| 156 |
+
"time_embedding_type": "positional",
|
| 157 |
+
"timestep_post_act": null,
|
| 158 |
+
"transformer_layers_per_block": [
|
| 159 |
+
1,
|
| 160 |
+
2
|
| 161 |
+
],
|
| 162 |
+
"up_block_types": [
|
| 163 |
+
"CrossAttnUpBlock2D",
|
| 164 |
+
"UpBlock2D"
|
| 165 |
+
],
|
| 166 |
+
"upcast_attention": false,
|
| 167 |
+
"use_linear_projection": true
|
| 168 |
+
}
|
| 169 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/73e765061ca0edc0984c.json
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "MultiModelCacheEntry",
|
| 3 |
+
"_model_id": "/home/tengomucho/.cache/huggingface/hub/models--hf-internal-testing--tiny-stable-diffusion-torch/snapshots/a88cdfbd91f96ec7f61eb7484b652ff0f4ee701d/vae",
|
| 4 |
+
"_task": null,
|
| 5 |
+
"text_encoder": {
|
| 6 |
+
"architectures": [
|
| 7 |
+
"CLIPTextModel"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dropout": 0.0,
|
| 11 |
+
"dtype": "float32",
|
| 12 |
+
"export_model_type": "clip-text-model",
|
| 13 |
+
"hidden_act": "quick_gelu",
|
| 14 |
+
"hidden_size": 32,
|
| 15 |
+
"initializer_factor": 1.0,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 37,
|
| 18 |
+
"layer_norm_eps": 1e-05,
|
| 19 |
+
"max_position_embeddings": 77,
|
| 20 |
+
"model_type": "clip_text_model",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"auto_cast": "matmul",
|
| 23 |
+
"auto_cast_type": "bf16",
|
| 24 |
+
"compiler_type": "neuronx-cc",
|
| 25 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 26 |
+
"dynamic_batch_size": false,
|
| 27 |
+
"float_dtype": "fp32",
|
| 28 |
+
"inline_weights_to_neff": true,
|
| 29 |
+
"instance_type": "inf2",
|
| 30 |
+
"int_dtype": "int64",
|
| 31 |
+
"optlevel": "2",
|
| 32 |
+
"output_attentions": false,
|
| 33 |
+
"output_hidden_states": false,
|
| 34 |
+
"static_batch_size": 1,
|
| 35 |
+
"static_sequence_length": 77,
|
| 36 |
+
"task": "feature-extraction",
|
| 37 |
+
"tensor_parallel_size": 1
|
| 38 |
+
},
|
| 39 |
+
"num_attention_heads": 4,
|
| 40 |
+
"num_hidden_layers": 5,
|
| 41 |
+
"vocab_size": 1000
|
| 42 |
+
},
|
| 43 |
+
"unet": {
|
| 44 |
+
"_class_name": "UNet2DConditionModel",
|
| 45 |
+
"act_fn": "silu",
|
| 46 |
+
"addition_embed_type": null,
|
| 47 |
+
"addition_embed_type_num_heads": 64,
|
| 48 |
+
"addition_time_embed_dim": null,
|
| 49 |
+
"attention_head_dim": 8,
|
| 50 |
+
"attention_type": "default",
|
| 51 |
+
"block_out_channels": [
|
| 52 |
+
32,
|
| 53 |
+
64
|
| 54 |
+
],
|
| 55 |
+
"center_input_sample": false,
|
| 56 |
+
"class_embed_type": null,
|
| 57 |
+
"class_embeddings_concat": false,
|
| 58 |
+
"conv_in_kernel": 3,
|
| 59 |
+
"conv_out_kernel": 3,
|
| 60 |
+
"cross_attention_dim": 32,
|
| 61 |
+
"cross_attention_norm": null,
|
| 62 |
+
"down_block_types": [
|
| 63 |
+
"DownBlock2D",
|
| 64 |
+
"CrossAttnDownBlock2D"
|
| 65 |
+
],
|
| 66 |
+
"downsample_padding": 1,
|
| 67 |
+
"dropout": 0.0,
|
| 68 |
+
"dual_cross_attention": false,
|
| 69 |
+
"encoder_hid_dim": null,
|
| 70 |
+
"encoder_hid_dim_type": null,
|
| 71 |
+
"flip_sin_to_cos": true,
|
| 72 |
+
"freq_shift": 0,
|
| 73 |
+
"in_channels": 4,
|
| 74 |
+
"layers_per_block": 2,
|
| 75 |
+
"mid_block_only_cross_attention": null,
|
| 76 |
+
"mid_block_scale_factor": 1,
|
| 77 |
+
"mid_block_type": "UNetMidBlock2DCrossAttn",
|
| 78 |
+
"neuron": {
|
| 79 |
+
"auto_cast": "matmul",
|
| 80 |
+
"auto_cast_type": "bf16",
|
| 81 |
+
"compiler_type": "neuronx-cc",
|
| 82 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 83 |
+
"dynamic_batch_size": false,
|
| 84 |
+
"float_dtype": "fp32",
|
| 85 |
+
"inline_weights_to_neff": true,
|
| 86 |
+
"instance_type": "inf2",
|
| 87 |
+
"int_dtype": "int64",
|
| 88 |
+
"optlevel": "2",
|
| 89 |
+
"output_attentions": false,
|
| 90 |
+
"output_hidden_states": false,
|
| 91 |
+
"static_batch_size": 1,
|
| 92 |
+
"static_height": 32,
|
| 93 |
+
"static_num_channels": 4,
|
| 94 |
+
"static_sequence_length": 77,
|
| 95 |
+
"static_vae_scale_factor": 2,
|
| 96 |
+
"static_width": 32,
|
| 97 |
+
"task": "semantic-segmentation",
|
| 98 |
+
"tensor_parallel_size": 1
|
| 99 |
+
},
|
| 100 |
+
"norm_eps": 1e-05,
|
| 101 |
+
"norm_num_groups": 32,
|
| 102 |
+
"num_attention_heads": null,
|
| 103 |
+
"num_class_embeds": null,
|
| 104 |
+
"only_cross_attention": false,
|
| 105 |
+
"out_channels": 4,
|
| 106 |
+
"output_attentions": false,
|
| 107 |
+
"projection_class_embeddings_input_dim": null,
|
| 108 |
+
"resnet_out_scale_factor": 1.0,
|
| 109 |
+
"resnet_skip_time_act": false,
|
| 110 |
+
"resnet_time_scale_shift": "default",
|
| 111 |
+
"reverse_transformer_layers_per_block": null,
|
| 112 |
+
"time_cond_proj_dim": null,
|
| 113 |
+
"time_embedding_act_fn": null,
|
| 114 |
+
"time_embedding_dim": null,
|
| 115 |
+
"time_embedding_type": "positional",
|
| 116 |
+
"timestep_post_act": null,
|
| 117 |
+
"transformer_layers_per_block": 1,
|
| 118 |
+
"up_block_types": [
|
| 119 |
+
"CrossAttnUpBlock2D",
|
| 120 |
+
"UpBlock2D"
|
| 121 |
+
],
|
| 122 |
+
"upcast_attention": false,
|
| 123 |
+
"use_linear_projection": false
|
| 124 |
+
}
|
| 125 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/b191d607e94819566800de30939d6c26cefbebc52bef9ff571aa1855a9ca0a4e/c6310bffbfbf29e5d927.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "MultiModelCacheEntry",
|
| 3 |
+
"_model_id": "/home/tengomucho/.cache/huggingface/hub/models--echarlaix--tiny-random-latent-consistency/snapshots/ea16e1fcf89fe667eef6639bfd14ac8701bff7b9/vae",
|
| 4 |
+
"_task": null,
|
| 5 |
+
"text_encoder": {
|
| 6 |
+
"architectures": [
|
| 7 |
+
"CLIPTextModel"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"export_model_type": "clip-text-model",
|
| 12 |
+
"hidden_act": "quick_gelu",
|
| 13 |
+
"hidden_size": 32,
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 64,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 77,
|
| 19 |
+
"model_type": "clip_text_model",
|
| 20 |
+
"neuron": {
|
| 21 |
+
"auto_cast": "matmul",
|
| 22 |
+
"auto_cast_type": "bf16",
|
| 23 |
+
"compiler_type": "neuronx-cc",
|
| 24 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 25 |
+
"dynamic_batch_size": false,
|
| 26 |
+
"float_dtype": "fp32",
|
| 27 |
+
"inline_weights_to_neff": true,
|
| 28 |
+
"instance_type": "inf2",
|
| 29 |
+
"int_dtype": "int64",
|
| 30 |
+
"optlevel": "2",
|
| 31 |
+
"output_attentions": false,
|
| 32 |
+
"output_hidden_states": false,
|
| 33 |
+
"static_batch_size": 1,
|
| 34 |
+
"static_sequence_length": 77,
|
| 35 |
+
"task": "feature-extraction",
|
| 36 |
+
"tensor_parallel_size": 1
|
| 37 |
+
},
|
| 38 |
+
"num_attention_heads": 8,
|
| 39 |
+
"num_hidden_layers": 3,
|
| 40 |
+
"vocab_size": 1000
|
| 41 |
+
},
|
| 42 |
+
"unet": {
|
| 43 |
+
"_class_name": "UNet2DConditionModel",
|
| 44 |
+
"act_fn": "silu",
|
| 45 |
+
"addition_embed_type": null,
|
| 46 |
+
"addition_embed_type_num_heads": 64,
|
| 47 |
+
"addition_time_embed_dim": null,
|
| 48 |
+
"attention_head_dim": 8,
|
| 49 |
+
"attention_type": "default",
|
| 50 |
+
"block_out_channels": [
|
| 51 |
+
4,
|
| 52 |
+
8
|
| 53 |
+
],
|
| 54 |
+
"center_input_sample": false,
|
| 55 |
+
"class_embed_type": null,
|
| 56 |
+
"class_embeddings_concat": false,
|
| 57 |
+
"conv_in_kernel": 3,
|
| 58 |
+
"conv_out_kernel": 3,
|
| 59 |
+
"cross_attention_dim": 32,
|
| 60 |
+
"cross_attention_norm": null,
|
| 61 |
+
"down_block_types": [
|
| 62 |
+
"DownBlock2D",
|
| 63 |
+
"CrossAttnDownBlock2D"
|
| 64 |
+
],
|
| 65 |
+
"downsample_padding": 1,
|
| 66 |
+
"dropout": 0.0,
|
| 67 |
+
"dual_cross_attention": false,
|
| 68 |
+
"encoder_hid_dim": null,
|
| 69 |
+
"encoder_hid_dim_type": null,
|
| 70 |
+
"flip_sin_to_cos": true,
|
| 71 |
+
"freq_shift": 0,
|
| 72 |
+
"in_channels": 4,
|
| 73 |
+
"layers_per_block": 1,
|
| 74 |
+
"mid_block_only_cross_attention": null,
|
| 75 |
+
"mid_block_scale_factor": 1,
|
| 76 |
+
"mid_block_type": "UNetMidBlock2DCrossAttn",
|
| 77 |
+
"neuron": {
|
| 78 |
+
"auto_cast": "matmul",
|
| 79 |
+
"auto_cast_type": "bf16",
|
| 80 |
+
"compiler_type": "neuronx-cc",
|
| 81 |
+
"compiler_version": "2.26.6360.0+6f180f47",
|
| 82 |
+
"dynamic_batch_size": false,
|
| 83 |
+
"float_dtype": "fp32",
|
| 84 |
+
"inline_weights_to_neff": true,
|
| 85 |
+
"instance_type": "inf2",
|
| 86 |
+
"int_dtype": "int64",
|
| 87 |
+
"optlevel": "2",
|
| 88 |
+
"output_attentions": false,
|
| 89 |
+
"output_hidden_states": false,
|
| 90 |
+
"static_batch_size": 1,
|
| 91 |
+
"static_height": 32,
|
| 92 |
+
"static_num_channels": 4,
|
| 93 |
+
"static_sequence_length": 77,
|
| 94 |
+
"static_vae_scale_factor": 2,
|
| 95 |
+
"static_width": 32,
|
| 96 |
+
"task": "semantic-segmentation",
|
| 97 |
+
"tensor_parallel_size": 1
|
| 98 |
+
},
|
| 99 |
+
"norm_eps": 1e-05,
|
| 100 |
+
"norm_num_groups": 2,
|
| 101 |
+
"num_attention_heads": null,
|
| 102 |
+
"num_class_embeds": null,
|
| 103 |
+
"only_cross_attention": false,
|
| 104 |
+
"out_channels": 4,
|
| 105 |
+
"output_attentions": false,
|
| 106 |
+
"projection_class_embeddings_input_dim": null,
|
| 107 |
+
"resnet_out_scale_factor": 1.0,
|
| 108 |
+
"resnet_skip_time_act": false,
|
| 109 |
+
"resnet_time_scale_shift": "default",
|
| 110 |
+
"reverse_transformer_layers_per_block": null,
|
| 111 |
+
"time_cond_proj_dim": 32,
|
| 112 |
+
"time_embedding_act_fn": null,
|
| 113 |
+
"time_embedding_dim": null,
|
| 114 |
+
"time_embedding_type": "positional",
|
| 115 |
+
"timestep_post_act": null,
|
| 116 |
+
"transformer_layers_per_block": 1,
|
| 117 |
+
"up_block_types": [
|
| 118 |
+
"CrossAttnUpBlock2D",
|
| 119 |
+
"UpBlock2D"
|
| 120 |
+
],
|
| 121 |
+
"upcast_attention": false,
|
| 122 |
+
"use_linear_projection": false
|
| 123 |
+
}
|
| 124 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/190c48cba7eeccd29748.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": null,
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": false,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 4096,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 4096,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": false,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 4096,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "bfloat16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_hidden_layers": 16,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 32.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"unsloth_fixed": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 128256
|
| 64 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/4709e472b680eafc6cc9.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 8192,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 8192,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 1024,
|
| 40 |
+
"sequence_length": 8192,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "bfloat16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_hidden_layers": 16,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 32.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"unsloth_fixed": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 128256
|
| 64 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/552e8cb4a475fa6bf79e.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 4,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
|
| 25 |
+
"continuous_batching": true,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 4,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 1024,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "bfloat16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_hidden_layers": 16,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 32.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"unsloth_fixed": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 128256
|
| 64 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/cf6b9a360dcf294104671106bae2adbd9fd291823bb60a351883163684073231/d391df1950611ad2a02d.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": null,
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": false,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 4096,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 4096,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": false,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 4096,
|
| 41 |
+
"speculation_length": 5,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "bfloat16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_hidden_layers": 16,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 32.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"unsloth_fixed": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 128256
|
| 64 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/e0a07fcbaaaca6e450fabdb638553525f4bfb07bcac85320ca02556f31cd98cf/68b8b60ad74d23a1ba59.json
ADDED
|
@@ -0,0 +1,130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-4b-it",
|
| 4 |
+
"_task": "image-text-to-text",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Gemma3ForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"boi_token_index": 255999,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eoi_token_index": 256000,
|
| 11 |
+
"image_token_index": 262144,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"mm_tokens_per_image": 256,
|
| 14 |
+
"model_type": "gemma3",
|
| 15 |
+
"neuron": {
|
| 16 |
+
"_serialized_key": "NxDVLMNeuronConfig",
|
| 17 |
+
"batch_size": 2,
|
| 18 |
+
"capacity_factor": null,
|
| 19 |
+
"checkpoint_id": "unsloth/gemma-3-4b-it",
|
| 20 |
+
"checkpoint_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
|
| 21 |
+
"continuous_batching": true,
|
| 22 |
+
"ep_degree": 1,
|
| 23 |
+
"fused_qkv": true,
|
| 24 |
+
"glu_mlp": true,
|
| 25 |
+
"image_seq_len": 256,
|
| 26 |
+
"image_size": 896,
|
| 27 |
+
"local_ranks_size": 2,
|
| 28 |
+
"max_batch_size": 2,
|
| 29 |
+
"max_context_length": 2048,
|
| 30 |
+
"max_num_images": 5,
|
| 31 |
+
"max_topk": 256,
|
| 32 |
+
"n_active_tokens": 2048,
|
| 33 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 34 |
+
"on_device_sampling": true,
|
| 35 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 36 |
+
"output_logits": false,
|
| 37 |
+
"pp_degree": 1,
|
| 38 |
+
"prefill_chunk_size": 0,
|
| 39 |
+
"sequence_length": 2048,
|
| 40 |
+
"speculation_length": 0,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "bfloat16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"text_config": {
|
| 47 |
+
"_sliding_window_pattern": 6,
|
| 48 |
+
"attention_bias": false,
|
| 49 |
+
"attention_dropout": 0.0,
|
| 50 |
+
"attn_logit_softcapping": null,
|
| 51 |
+
"cache_implementation": "hybrid",
|
| 52 |
+
"dtype": "bfloat16",
|
| 53 |
+
"final_logit_softcapping": null,
|
| 54 |
+
"head_dim": 256,
|
| 55 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 56 |
+
"hidden_size": 2560,
|
| 57 |
+
"initializer_range": 0.02,
|
| 58 |
+
"intermediate_size": 10240,
|
| 59 |
+
"layer_types": [
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"sliding_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"full_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"sliding_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"full_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"sliding_attention",
|
| 76 |
+
"sliding_attention",
|
| 77 |
+
"full_attention",
|
| 78 |
+
"sliding_attention",
|
| 79 |
+
"sliding_attention",
|
| 80 |
+
"sliding_attention",
|
| 81 |
+
"sliding_attention",
|
| 82 |
+
"sliding_attention",
|
| 83 |
+
"full_attention",
|
| 84 |
+
"sliding_attention",
|
| 85 |
+
"sliding_attention",
|
| 86 |
+
"sliding_attention",
|
| 87 |
+
"sliding_attention",
|
| 88 |
+
"sliding_attention",
|
| 89 |
+
"full_attention",
|
| 90 |
+
"sliding_attention",
|
| 91 |
+
"sliding_attention",
|
| 92 |
+
"sliding_attention",
|
| 93 |
+
"sliding_attention"
|
| 94 |
+
],
|
| 95 |
+
"max_position_embeddings": 131072,
|
| 96 |
+
"model_type": "gemma3_text",
|
| 97 |
+
"num_attention_heads": 8,
|
| 98 |
+
"num_hidden_layers": 34,
|
| 99 |
+
"num_key_value_heads": 4,
|
| 100 |
+
"query_pre_attn_scalar": 256,
|
| 101 |
+
"rms_norm_eps": 1e-06,
|
| 102 |
+
"rope_local_base_freq": 10000.0,
|
| 103 |
+
"rope_scaling": {
|
| 104 |
+
"factor": 8.0,
|
| 105 |
+
"rope_type": "linear"
|
| 106 |
+
},
|
| 107 |
+
"rope_theta": 1000000.0,
|
| 108 |
+
"sliding_window": 1024,
|
| 109 |
+
"sliding_window_pattern": 6,
|
| 110 |
+
"use_bidirectional_attention": false,
|
| 111 |
+
"use_cache": true,
|
| 112 |
+
"vocab_size": 262208
|
| 113 |
+
},
|
| 114 |
+
"unsloth_fixed": true,
|
| 115 |
+
"vision_config": {
|
| 116 |
+
"attention_dropout": 0.0,
|
| 117 |
+
"dtype": "bfloat16",
|
| 118 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 119 |
+
"hidden_size": 1152,
|
| 120 |
+
"image_size": 896,
|
| 121 |
+
"intermediate_size": 4304,
|
| 122 |
+
"layer_norm_eps": 1e-06,
|
| 123 |
+
"model_type": "siglip_vision_model",
|
| 124 |
+
"num_attention_heads": 16,
|
| 125 |
+
"num_channels": 3,
|
| 126 |
+
"num_hidden_layers": 27,
|
| 127 |
+
"patch_size": 14,
|
| 128 |
+
"vision_use_head": false
|
| 129 |
+
}
|
| 130 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3/unsloth/gemma-3-4b-it/68b8b60ad74d23a1ba59.json
ADDED
|
@@ -0,0 +1,130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-4b-it",
|
| 4 |
+
"_task": "image-text-to-text",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Gemma3ForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"boi_token_index": 255999,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eoi_token_index": 256000,
|
| 11 |
+
"image_token_index": 262144,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"mm_tokens_per_image": 256,
|
| 14 |
+
"model_type": "gemma3",
|
| 15 |
+
"neuron": {
|
| 16 |
+
"_serialized_key": "NxDVLMNeuronConfig",
|
| 17 |
+
"batch_size": 2,
|
| 18 |
+
"capacity_factor": null,
|
| 19 |
+
"checkpoint_id": "unsloth/gemma-3-4b-it",
|
| 20 |
+
"checkpoint_revision": "bf46152c47f5dd20b896357cb51abc4c03b8ee8c",
|
| 21 |
+
"continuous_batching": true,
|
| 22 |
+
"ep_degree": 1,
|
| 23 |
+
"fused_qkv": true,
|
| 24 |
+
"glu_mlp": true,
|
| 25 |
+
"image_seq_len": 256,
|
| 26 |
+
"image_size": 896,
|
| 27 |
+
"local_ranks_size": 2,
|
| 28 |
+
"max_batch_size": 2,
|
| 29 |
+
"max_context_length": 2048,
|
| 30 |
+
"max_num_images": 5,
|
| 31 |
+
"max_topk": 256,
|
| 32 |
+
"n_active_tokens": 2048,
|
| 33 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 34 |
+
"on_device_sampling": true,
|
| 35 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 36 |
+
"output_logits": false,
|
| 37 |
+
"pp_degree": 1,
|
| 38 |
+
"prefill_chunk_size": 0,
|
| 39 |
+
"sequence_length": 2048,
|
| 40 |
+
"speculation_length": 0,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "bfloat16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"text_config": {
|
| 47 |
+
"_sliding_window_pattern": 6,
|
| 48 |
+
"attention_bias": false,
|
| 49 |
+
"attention_dropout": 0.0,
|
| 50 |
+
"attn_logit_softcapping": null,
|
| 51 |
+
"cache_implementation": "hybrid",
|
| 52 |
+
"dtype": "bfloat16",
|
| 53 |
+
"final_logit_softcapping": null,
|
| 54 |
+
"head_dim": 256,
|
| 55 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 56 |
+
"hidden_size": 2560,
|
| 57 |
+
"initializer_range": 0.02,
|
| 58 |
+
"intermediate_size": 10240,
|
| 59 |
+
"layer_types": [
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"sliding_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"full_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"sliding_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"full_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"sliding_attention",
|
| 76 |
+
"sliding_attention",
|
| 77 |
+
"full_attention",
|
| 78 |
+
"sliding_attention",
|
| 79 |
+
"sliding_attention",
|
| 80 |
+
"sliding_attention",
|
| 81 |
+
"sliding_attention",
|
| 82 |
+
"sliding_attention",
|
| 83 |
+
"full_attention",
|
| 84 |
+
"sliding_attention",
|
| 85 |
+
"sliding_attention",
|
| 86 |
+
"sliding_attention",
|
| 87 |
+
"sliding_attention",
|
| 88 |
+
"sliding_attention",
|
| 89 |
+
"full_attention",
|
| 90 |
+
"sliding_attention",
|
| 91 |
+
"sliding_attention",
|
| 92 |
+
"sliding_attention",
|
| 93 |
+
"sliding_attention"
|
| 94 |
+
],
|
| 95 |
+
"max_position_embeddings": 131072,
|
| 96 |
+
"model_type": "gemma3_text",
|
| 97 |
+
"num_attention_heads": 8,
|
| 98 |
+
"num_hidden_layers": 34,
|
| 99 |
+
"num_key_value_heads": 4,
|
| 100 |
+
"query_pre_attn_scalar": 256,
|
| 101 |
+
"rms_norm_eps": 1e-06,
|
| 102 |
+
"rope_local_base_freq": 10000.0,
|
| 103 |
+
"rope_scaling": {
|
| 104 |
+
"factor": 8.0,
|
| 105 |
+
"rope_type": "linear"
|
| 106 |
+
},
|
| 107 |
+
"rope_theta": 1000000.0,
|
| 108 |
+
"sliding_window": 1024,
|
| 109 |
+
"sliding_window_pattern": 6,
|
| 110 |
+
"use_bidirectional_attention": false,
|
| 111 |
+
"use_cache": true,
|
| 112 |
+
"vocab_size": 262208
|
| 113 |
+
},
|
| 114 |
+
"unsloth_fixed": true,
|
| 115 |
+
"vision_config": {
|
| 116 |
+
"attention_dropout": 0.0,
|
| 117 |
+
"dtype": "bfloat16",
|
| 118 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 119 |
+
"hidden_size": 1152,
|
| 120 |
+
"image_size": 896,
|
| 121 |
+
"intermediate_size": 4304,
|
| 122 |
+
"layer_norm_eps": 1e-06,
|
| 123 |
+
"model_type": "siglip_vision_model",
|
| 124 |
+
"num_attention_heads": 16,
|
| 125 |
+
"num_channels": 3,
|
| 126 |
+
"num_hidden_layers": 27,
|
| 127 |
+
"patch_size": 14,
|
| 128 |
+
"vision_use_head": false
|
| 129 |
+
}
|
| 130 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3_text/unsloth/gemma-3-270m-it/748542be1bcc5fe3cdf1.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/gemma-3-270m-it",
|
| 4 |
+
"_sliding_window_pattern": 6,
|
| 5 |
+
"_task": "text-generation",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"Gemma3ForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"attn_logit_softcapping": null,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"final_logit_softcapping": null,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 640,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 2048,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"sliding_attention",
|
| 21 |
+
"sliding_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"sliding_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"sliding_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"sliding_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"full_attention"
|
| 38 |
+
],
|
| 39 |
+
"max_position_embeddings": 32768,
|
| 40 |
+
"model_type": "gemma3_text",
|
| 41 |
+
"neuron": {
|
| 42 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 43 |
+
"batch_size": 1,
|
| 44 |
+
"capacity_factor": null,
|
| 45 |
+
"checkpoint_id": "unsloth/gemma-3-270m-it",
|
| 46 |
+
"checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
|
| 47 |
+
"continuous_batching": false,
|
| 48 |
+
"ep_degree": 1,
|
| 49 |
+
"fused_qkv": true,
|
| 50 |
+
"glu_mlp": true,
|
| 51 |
+
"local_ranks_size": 2,
|
| 52 |
+
"max_batch_size": 1,
|
| 53 |
+
"max_context_length": 8192,
|
| 54 |
+
"max_topk": 256,
|
| 55 |
+
"n_active_tokens": 8192,
|
| 56 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 57 |
+
"on_device_sampling": true,
|
| 58 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 59 |
+
"output_logits": false,
|
| 60 |
+
"pp_degree": 1,
|
| 61 |
+
"prefill_chunk_size": 0,
|
| 62 |
+
"sequence_length": 8192,
|
| 63 |
+
"speculation_length": 0,
|
| 64 |
+
"start_rank_id": 0,
|
| 65 |
+
"target": "trn1",
|
| 66 |
+
"torch_dtype": "bfloat16",
|
| 67 |
+
"tp_degree": 2
|
| 68 |
+
},
|
| 69 |
+
"num_attention_heads": 4,
|
| 70 |
+
"num_hidden_layers": 18,
|
| 71 |
+
"num_key_value_heads": 1,
|
| 72 |
+
"query_pre_attn_scalar": 256,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_local_base_freq": 10000.0,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": 512,
|
| 78 |
+
"unsloth_fixed": true,
|
| 79 |
+
"use_bidirectional_attention": false,
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 262144
|
| 82 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/granite/ibm-granite/granite-3.1-2b-instruct/2132439b615164bef4eb.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 12 |
+
"embedding_multiplier": 12.0,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 2048,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 8192,
|
| 17 |
+
"logits_scaling": 8.0,
|
| 18 |
+
"max_position_embeddings": 131072,
|
| 19 |
+
"mlp_bias": false,
|
| 20 |
+
"model_type": "granite",
|
| 21 |
+
"neuron": {
|
| 22 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 23 |
+
"batch_size": 4,
|
| 24 |
+
"capacity_factor": null,
|
| 25 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
| 26 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
| 27 |
+
"continuous_batching": true,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": true,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 2,
|
| 32 |
+
"max_batch_size": 4,
|
| 33 |
+
"max_context_length": 1024,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 1024,
|
| 36 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"prefill_chunk_size": 0,
|
| 42 |
+
"sequence_length": 1024,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": "trn1",
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"tp_degree": 2
|
| 48 |
+
},
|
| 49 |
+
"num_attention_heads": 32,
|
| 50 |
+
"num_hidden_layers": 40,
|
| 51 |
+
"num_key_value_heads": 8,
|
| 52 |
+
"residual_multiplier": 0.22,
|
| 53 |
+
"rms_norm_eps": 1e-05,
|
| 54 |
+
"rope_scaling": null,
|
| 55 |
+
"rope_theta": 5000000.0,
|
| 56 |
+
"tie_word_embeddings": true,
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 49155
|
| 59 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/idefics3/HuggingFaceTB/SmolVLM-256M-Instruct/ec1560a24138c907085c.json
ADDED
|
@@ -0,0 +1,189 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolVLM-256M-Instruct",
|
| 4 |
+
"_task": "image-text-to-text",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Idefics3ForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"image_token_id": 49190,
|
| 10 |
+
"model_type": "idefics3",
|
| 11 |
+
"neuron": {
|
| 12 |
+
"_serialized_key": "NxDVLMNeuronConfig",
|
| 13 |
+
"batch_size": 2,
|
| 14 |
+
"capacity_factor": null,
|
| 15 |
+
"checkpoint_id": "HuggingFaceTB/SmolVLM-256M-Instruct",
|
| 16 |
+
"checkpoint_revision": "7e3e67edbbed1bf9888184d9df282b700a323964",
|
| 17 |
+
"continuous_batching": true,
|
| 18 |
+
"ep_degree": 1,
|
| 19 |
+
"fused_qkv": true,
|
| 20 |
+
"glu_mlp": true,
|
| 21 |
+
"image_seq_len": 64,
|
| 22 |
+
"image_size": 512,
|
| 23 |
+
"local_ranks_size": 2,
|
| 24 |
+
"max_batch_size": 2,
|
| 25 |
+
"max_context_length": 2048,
|
| 26 |
+
"max_num_images": 17,
|
| 27 |
+
"max_topk": 256,
|
| 28 |
+
"n_active_tokens": 2048,
|
| 29 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 30 |
+
"on_device_sampling": true,
|
| 31 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 32 |
+
"output_logits": false,
|
| 33 |
+
"pp_degree": 1,
|
| 34 |
+
"prefill_chunk_size": 1024,
|
| 35 |
+
"sequence_length": 2048,
|
| 36 |
+
"speculation_length": 0,
|
| 37 |
+
"start_rank_id": 0,
|
| 38 |
+
"target": "trn1",
|
| 39 |
+
"torch_dtype": "bfloat16",
|
| 40 |
+
"tp_degree": 2
|
| 41 |
+
},
|
| 42 |
+
"scale_factor": 4,
|
| 43 |
+
"text_config": {
|
| 44 |
+
"_attn_implementation_autoset": false,
|
| 45 |
+
"_flash_attn_2_enabled": true,
|
| 46 |
+
"_name_or_path": "None",
|
| 47 |
+
"architectures": [
|
| 48 |
+
"VLlama3ForCausalLM"
|
| 49 |
+
],
|
| 50 |
+
"attention_bias": false,
|
| 51 |
+
"attention_dropout": 0.0,
|
| 52 |
+
"dtype": "bfloat16",
|
| 53 |
+
"head_dim": 64,
|
| 54 |
+
"hidden_act": "silu",
|
| 55 |
+
"hidden_size": 576,
|
| 56 |
+
"initializer_range": 0.041666666666666664,
|
| 57 |
+
"intermediate_size": 1536,
|
| 58 |
+
"is_llama_config": true,
|
| 59 |
+
"max_position_embeddings": 8192,
|
| 60 |
+
"mlp_bias": false,
|
| 61 |
+
"model_type": "llama",
|
| 62 |
+
"neftune_noise_alpha": 0.0,
|
| 63 |
+
"num_attention_heads": 9,
|
| 64 |
+
"num_hidden_layers": 30,
|
| 65 |
+
"num_key_value_heads": 3,
|
| 66 |
+
"pad_token_id": 2,
|
| 67 |
+
"perceiver_config": {
|
| 68 |
+
"_attn_implementation_autoset": false,
|
| 69 |
+
"_name_or_path": "",
|
| 70 |
+
"add_cross_attention": false,
|
| 71 |
+
"architectures": null,
|
| 72 |
+
"attention_dropout": 0.0,
|
| 73 |
+
"bad_words_ids": null,
|
| 74 |
+
"begin_suppress_tokens": null,
|
| 75 |
+
"bos_token_id": null,
|
| 76 |
+
"chunk_size_feed_forward": 0,
|
| 77 |
+
"cross_attention_hidden_size": null,
|
| 78 |
+
"decoder_start_token_id": null,
|
| 79 |
+
"diversity_penalty": 0.0,
|
| 80 |
+
"do_sample": false,
|
| 81 |
+
"early_stopping": false,
|
| 82 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 83 |
+
"eos_token_id": null,
|
| 84 |
+
"exponential_decay_length_penalty": null,
|
| 85 |
+
"finetuning_task": null,
|
| 86 |
+
"forced_bos_token_id": null,
|
| 87 |
+
"forced_eos_token_id": null,
|
| 88 |
+
"hidden_act": "silu",
|
| 89 |
+
"id2label": {
|
| 90 |
+
"0": "LABEL_0",
|
| 91 |
+
"1": "LABEL_1"
|
| 92 |
+
},
|
| 93 |
+
"is_decoder": false,
|
| 94 |
+
"is_encoder_decoder": false,
|
| 95 |
+
"label2id": {
|
| 96 |
+
"LABEL_0": 0,
|
| 97 |
+
"LABEL_1": 1
|
| 98 |
+
},
|
| 99 |
+
"length_penalty": 1.0,
|
| 100 |
+
"max_length": 20,
|
| 101 |
+
"min_length": 0,
|
| 102 |
+
"model_type": "vllama3",
|
| 103 |
+
"no_repeat_ngram_size": 0,
|
| 104 |
+
"num_beam_groups": 1,
|
| 105 |
+
"num_beams": 1,
|
| 106 |
+
"num_key_value_heads": 1,
|
| 107 |
+
"num_return_sequences": 1,
|
| 108 |
+
"output_attentions": false,
|
| 109 |
+
"output_hidden_states": false,
|
| 110 |
+
"output_scores": false,
|
| 111 |
+
"pad_token_id": null,
|
| 112 |
+
"prefix": null,
|
| 113 |
+
"problem_type": null,
|
| 114 |
+
"pruned_heads": {},
|
| 115 |
+
"qk_layer_norms_perceiver": false,
|
| 116 |
+
"remove_invalid_values": false,
|
| 117 |
+
"repetition_penalty": 1.0,
|
| 118 |
+
"resampler_depth": 6,
|
| 119 |
+
"resampler_head_dim": 96,
|
| 120 |
+
"resampler_n_heads": 16,
|
| 121 |
+
"resampler_n_latents": 64,
|
| 122 |
+
"return_dict": true,
|
| 123 |
+
"return_dict_in_generate": false,
|
| 124 |
+
"sep_token_id": null,
|
| 125 |
+
"suppress_tokens": null,
|
| 126 |
+
"task_specific_params": null,
|
| 127 |
+
"temperature": 1.0,
|
| 128 |
+
"tf_legacy_loss": false,
|
| 129 |
+
"tie_encoder_decoder": false,
|
| 130 |
+
"tie_word_embeddings": true,
|
| 131 |
+
"tokenizer_class": null,
|
| 132 |
+
"top_k": 50,
|
| 133 |
+
"top_p": 1.0,
|
| 134 |
+
"torch_dtype": null,
|
| 135 |
+
"torchscript": false,
|
| 136 |
+
"transformers_version": "4.46.0",
|
| 137 |
+
"typical_p": 1.0,
|
| 138 |
+
"use_bfloat16": false
|
| 139 |
+
},
|
| 140 |
+
"pixel_shuffle_factor": 4,
|
| 141 |
+
"pretraining_tp": 1,
|
| 142 |
+
"qk_layer_norms": false,
|
| 143 |
+
"rms_norm_eps": 1e-05,
|
| 144 |
+
"rope_interleaved": false,
|
| 145 |
+
"rope_scaling": null,
|
| 146 |
+
"rope_theta": 100000,
|
| 147 |
+
"transformers.js_config": {
|
| 148 |
+
"kv_cache_dtype": {
|
| 149 |
+
"fp16": "float16",
|
| 150 |
+
"q4f16": "float16"
|
| 151 |
+
}
|
| 152 |
+
},
|
| 153 |
+
"use_cache": true,
|
| 154 |
+
"use_resampler": false,
|
| 155 |
+
"vocab_size": 49280
|
| 156 |
+
},
|
| 157 |
+
"tie_word_embeddings": false,
|
| 158 |
+
"transformers.js_config": {
|
| 159 |
+
"kv_cache_dtype": {
|
| 160 |
+
"fp16": "float16",
|
| 161 |
+
"q4f16": "float16"
|
| 162 |
+
}
|
| 163 |
+
},
|
| 164 |
+
"use_cache": true,
|
| 165 |
+
"vision_config": {
|
| 166 |
+
"_attn_implementation_autoset": false,
|
| 167 |
+
"attention_dropout": 0.0,
|
| 168 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 169 |
+
"hidden_size": 768,
|
| 170 |
+
"image_size": 512,
|
| 171 |
+
"initializer_range": 0.02,
|
| 172 |
+
"intermediate_size": 3072,
|
| 173 |
+
"layer_norm_eps": 1e-06,
|
| 174 |
+
"max_image_size": {
|
| 175 |
+
"longest_edge": 512
|
| 176 |
+
},
|
| 177 |
+
"model_type": "idefics3_vision",
|
| 178 |
+
"num_attention_heads": 12,
|
| 179 |
+
"num_channels": 3,
|
| 180 |
+
"num_hidden_layers": 12,
|
| 181 |
+
"patch_size": 16,
|
| 182 |
+
"size": {
|
| 183 |
+
"longest_edge": 2048
|
| 184 |
+
},
|
| 185 |
+
"tie_word_embeddings": false,
|
| 186 |
+
"use_base_siglip": true
|
| 187 |
+
},
|
| 188 |
+
"vocab_size": 49280
|
| 189 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/llamafactory/tiny-random-Llama-3/24892d985c7b4406e3d9.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-Llama-3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "float16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "llamafactory/tiny-random-Llama-3",
|
| 24 |
+
"checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 1024,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "float16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 4,
|
| 48 |
+
"num_hidden_layers": 2,
|
| 49 |
+
"num_key_value_heads": 4,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 8.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": false,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/unsloth/Llama-3.2-1B-Instruct/552e8cb4a475fa6bf79e.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 4,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
|
| 25 |
+
"continuous_batching": true,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 4,
|
| 31 |
+
"max_context_length": 1024,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 1024,
|
| 34 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"prefill_chunk_size": 0,
|
| 40 |
+
"sequence_length": 1024,
|
| 41 |
+
"speculation_length": 0,
|
| 42 |
+
"start_rank_id": 0,
|
| 43 |
+
"target": "trn1",
|
| 44 |
+
"torch_dtype": "bfloat16",
|
| 45 |
+
"tp_degree": 2
|
| 46 |
+
},
|
| 47 |
+
"num_attention_heads": 32,
|
| 48 |
+
"num_hidden_layers": 16,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pretraining_tp": 1,
|
| 51 |
+
"rms_norm_eps": 1e-05,
|
| 52 |
+
"rope_scaling": {
|
| 53 |
+
"factor": 32.0,
|
| 54 |
+
"high_freq_factor": 4.0,
|
| 55 |
+
"low_freq_factor": 1.0,
|
| 56 |
+
"original_max_position_embeddings": 8192,
|
| 57 |
+
"rope_type": "llama3"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 500000.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"unsloth_fixed": true,
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 128256
|
| 64 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen2/Qwen/Qwen2.5-0.5B/3474ddeda6545f7e5db7.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 896,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 4864,
|
| 14 |
+
"layer_types": [
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention"
|
| 39 |
+
],
|
| 40 |
+
"max_position_embeddings": 32768,
|
| 41 |
+
"max_window_layers": 24,
|
| 42 |
+
"model_type": "qwen2",
|
| 43 |
+
"neuron": {
|
| 44 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 45 |
+
"batch_size": 4,
|
| 46 |
+
"capacity_factor": null,
|
| 47 |
+
"checkpoint_id": "Qwen/Qwen2.5-0.5B",
|
| 48 |
+
"checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
|
| 49 |
+
"continuous_batching": true,
|
| 50 |
+
"ep_degree": 1,
|
| 51 |
+
"fused_qkv": false,
|
| 52 |
+
"glu_mlp": true,
|
| 53 |
+
"local_ranks_size": 2,
|
| 54 |
+
"max_batch_size": 4,
|
| 55 |
+
"max_context_length": 1024,
|
| 56 |
+
"max_topk": 256,
|
| 57 |
+
"n_active_tokens": 1024,
|
| 58 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 59 |
+
"on_device_sampling": false,
|
| 60 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 61 |
+
"output_logits": false,
|
| 62 |
+
"pp_degree": 1,
|
| 63 |
+
"prefill_chunk_size": 0,
|
| 64 |
+
"sequence_length": 1024,
|
| 65 |
+
"speculation_length": 0,
|
| 66 |
+
"start_rank_id": 0,
|
| 67 |
+
"target": "trn1",
|
| 68 |
+
"torch_dtype": "bfloat16",
|
| 69 |
+
"tp_degree": 2
|
| 70 |
+
},
|
| 71 |
+
"num_attention_heads": 14,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"num_key_value_heads": 2,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_scaling": null,
|
| 76 |
+
"rope_theta": 1000000.0,
|
| 77 |
+
"sliding_window": null,
|
| 78 |
+
"tie_word_embeddings": true,
|
| 79 |
+
"use_cache": true,
|
| 80 |
+
"use_mrope": false,
|
| 81 |
+
"use_sliding_window": false,
|
| 82 |
+
"vocab_size": 151936
|
| 83 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-0.6B/0b70cb2cd9d99c1bdf34.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 40960,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 4,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": true,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 1024,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 1024,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 0,
|
| 70 |
+
"sequence_length": 1024,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-Embedding-0.6B/61ff8eae2078e66cabfd.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 128,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention"
|
| 45 |
+
],
|
| 46 |
+
"max_position_embeddings": 32768,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 4,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "97b0c614be4d77ee51c0cef4e5f07c00f9eb65b3",
|
| 55 |
+
"continuous_batching": false,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 1024,
|
| 70 |
+
"sequence_length": 8192,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn1",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 2
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151669
|
| 88 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/llamafactory/tiny-random-qwen3/c56a8bce0577ce31a594.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "llamafactory/tiny-random-qwen3",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 4,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 16,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 64,
|
| 16 |
+
"layer_types": [
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention"
|
| 19 |
+
],
|
| 20 |
+
"max_position_embeddings": 32768,
|
| 21 |
+
"max_window_layers": 36,
|
| 22 |
+
"model_type": "qwen3",
|
| 23 |
+
"neuron": {
|
| 24 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 25 |
+
"batch_size": 2,
|
| 26 |
+
"capacity_factor": null,
|
| 27 |
+
"checkpoint_id": "llamafactory/tiny-random-qwen3",
|
| 28 |
+
"checkpoint_revision": "81d6f5f5e05ed53ea8a1d19431266a486e46bbd8",
|
| 29 |
+
"continuous_batching": true,
|
| 30 |
+
"ep_degree": 1,
|
| 31 |
+
"fused_qkv": true,
|
| 32 |
+
"glu_mlp": true,
|
| 33 |
+
"local_ranks_size": 2,
|
| 34 |
+
"max_batch_size": 2,
|
| 35 |
+
"max_context_length": 1024,
|
| 36 |
+
"max_topk": 256,
|
| 37 |
+
"n_active_tokens": 1024,
|
| 38 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 39 |
+
"on_device_sampling": false,
|
| 40 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 41 |
+
"output_logits": false,
|
| 42 |
+
"pp_degree": 1,
|
| 43 |
+
"prefill_chunk_size": 0,
|
| 44 |
+
"sequence_length": 1024,
|
| 45 |
+
"speculation_length": 0,
|
| 46 |
+
"start_rank_id": 0,
|
| 47 |
+
"target": "trn1",
|
| 48 |
+
"torch_dtype": "bfloat16",
|
| 49 |
+
"tp_degree": 2
|
| 50 |
+
},
|
| 51 |
+
"num_attention_heads": 4,
|
| 52 |
+
"num_hidden_layers": 2,
|
| 53 |
+
"num_key_value_heads": 4,
|
| 54 |
+
"rms_norm_eps": 1e-06,
|
| 55 |
+
"rope_scaling": null,
|
| 56 |
+
"rope_theta": 5000000,
|
| 57 |
+
"sliding_window": null,
|
| 58 |
+
"tie_word_embeddings": true,
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"use_sliding_window": false,
|
| 61 |
+
"vocab_size": 151936
|
| 62 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/smollm3/HuggingFaceTB/SmolLM3-3B/8d67a5db932b99602641.json
ADDED
|
@@ -0,0 +1,135 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolLM3-3B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"SmolLM3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 65536,
|
| 54 |
+
"max_window_layers": 28,
|
| 55 |
+
"mlp_bias": false,
|
| 56 |
+
"model_type": "smollm3",
|
| 57 |
+
"neuron": {
|
| 58 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 59 |
+
"batch_size": 4,
|
| 60 |
+
"capacity_factor": null,
|
| 61 |
+
"checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
|
| 62 |
+
"checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
|
| 63 |
+
"continuous_batching": true,
|
| 64 |
+
"ep_degree": 1,
|
| 65 |
+
"fused_qkv": true,
|
| 66 |
+
"glu_mlp": true,
|
| 67 |
+
"local_ranks_size": 2,
|
| 68 |
+
"max_batch_size": 4,
|
| 69 |
+
"max_context_length": 1024,
|
| 70 |
+
"max_topk": 256,
|
| 71 |
+
"n_active_tokens": 1024,
|
| 72 |
+
"neuronxcc_version": "2.26.6360.0+6f180f47",
|
| 73 |
+
"on_device_sampling": true,
|
| 74 |
+
"optimum_neuron_version": "0.4.7.dev0",
|
| 75 |
+
"output_logits": false,
|
| 76 |
+
"pp_degree": 1,
|
| 77 |
+
"prefill_chunk_size": 0,
|
| 78 |
+
"sequence_length": 1024,
|
| 79 |
+
"speculation_length": 0,
|
| 80 |
+
"start_rank_id": 0,
|
| 81 |
+
"target": "trn1",
|
| 82 |
+
"torch_dtype": "bfloat16",
|
| 83 |
+
"tp_degree": 2
|
| 84 |
+
},
|
| 85 |
+
"no_rope_layer_interval": 4,
|
| 86 |
+
"no_rope_layers": [
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
1,
|
| 90 |
+
0,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
1,
|
| 94 |
+
0,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
1,
|
| 98 |
+
0,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
1,
|
| 102 |
+
0,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
1,
|
| 106 |
+
0,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
1,
|
| 110 |
+
0,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
1,
|
| 114 |
+
0,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
1,
|
| 118 |
+
0,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
1,
|
| 122 |
+
0
|
| 123 |
+
],
|
| 124 |
+
"num_attention_heads": 16,
|
| 125 |
+
"num_hidden_layers": 36,
|
| 126 |
+
"num_key_value_heads": 4,
|
| 127 |
+
"pretraining_tp": 2,
|
| 128 |
+
"rms_norm_eps": 1e-06,
|
| 129 |
+
"rope_scaling": null,
|
| 130 |
+
"rope_theta": 5000000.0,
|
| 131 |
+
"sliding_window": null,
|
| 132 |
+
"use_cache": false,
|
| 133 |
+
"use_sliding_window": false,
|
| 134 |
+
"vocab_size": 128256
|
| 135 |
+
}
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--verbose=35", "--logfile=/tmp/nxd_model/encoding/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.done
ADDED
|
File without changes
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df61ceb956b65ffc93d7765ba86a6aacb1b771a8bacfddb7fa06002ebb021f6c
|
| 3 |
+
size 1691021
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_008268ac9b472a8ee1b2+97496b47/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e49a709e1ea870cad2e693b9c13b9e9a07ce0061b0a8ae3e8f132023e91218b9
|
| 3 |
+
size 26420224
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.done
ADDED
|
File without changes
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f23b364d4879cb413e0fbdd2dfb4dfeaa0821888feced3735fcfe5031b9d03d
|
| 3 |
+
size 2432816
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61b85ab2243de6c32a2756c56e7bf0171f30edb11c4024c6029ab9e66629ba0b
|
| 3 |
+
size 2653184
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0568499afedbce2755d2+84337dd9/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d3af7468affc34917a69e6713422ce841d8e7f7234ac3e1ed3a455b4f67a1e8
|
| 3 |
+
size 2813329
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0632e5fdcca8738ae989+12dd9119/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.26.6360.0+6f180f47/MODULE_0632e5fdcca8738ae989+12dd9119/model.done
ADDED
|
File without changes
|