Synchronizing local compiler cache.
Browse files- .gitattributes +1 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/gemma3_text/unsloth/gemma-3-270m-it/70022928da5b1a3a3562.json +81 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.neff +0 -0
.gitattributes
CHANGED
|
@@ -7044,3 +7044,4 @@ neuronxcc-2.21.33363.0+82129205/MODULE_c69743a3b026cb8c0f9e+a02c3a36/wrapped_nef
|
|
| 7044 |
neuronxcc-2.21.33363.0+82129205/MODULE_60f37ef2cb181ebeb434+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7045 |
neuronxcc-2.21.33363.0+82129205/MODULE_60f37ef2cb181ebeb434+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7046 |
neuronxcc-2.21.33363.0+82129205/MODULE_a36debd95d53c8bebd53+f7cce17f/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 7044 |
neuronxcc-2.21.33363.0+82129205/MODULE_60f37ef2cb181ebeb434+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7045 |
neuronxcc-2.21.33363.0+82129205/MODULE_60f37ef2cb181ebeb434+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7046 |
neuronxcc-2.21.33363.0+82129205/MODULE_a36debd95d53c8bebd53+f7cce17f/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7047 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/gemma3_text/unsloth/gemma-3-270m-it/70022928da5b1a3a3562.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "2.21.33363.0+82129205",
|
| 57 |
+
"on_device_sampling": true,
|
| 58 |
+
"optimum_neuron_version": "0.4.6.dev2",
|
| 59 |
+
"output_logits": false,
|
| 60 |
+
"pp_degree": 1,
|
| 61 |
+
"sequence_length": 8192,
|
| 62 |
+
"speculation_length": 0,
|
| 63 |
+
"start_rank_id": 0,
|
| 64 |
+
"target": "trn1",
|
| 65 |
+
"torch_dtype": "bfloat16",
|
| 66 |
+
"tp_degree": 2
|
| 67 |
+
},
|
| 68 |
+
"num_attention_heads": 4,
|
| 69 |
+
"num_hidden_layers": 18,
|
| 70 |
+
"num_key_value_heads": 1,
|
| 71 |
+
"query_pre_attn_scalar": 256,
|
| 72 |
+
"rms_norm_eps": 1e-06,
|
| 73 |
+
"rope_local_base_freq": 10000.0,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": 512,
|
| 77 |
+
"unsloth_fixed": true,
|
| 78 |
+
"use_bidirectional_attention": false,
|
| 79 |
+
"use_cache": true,
|
| 80 |
+
"vocab_size": 262144
|
| 81 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_b4137ebb-97b3-4dfb-b478-6ff75b21ebe0/compiler_workdir/OldSoftmaxModule/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:931bab94e0ad026baae4dc1e3e73ade90e2954dbeb86887622692d40ecb80fe9
|
| 3 |
+
size 4706
|
neuronxcc-2.21.33363.0+82129205/MODULE_1477804b8d4dee16534e+5047b4cf/model.neff
ADDED
|
Binary file (42 kB). View file
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_1cc78844-e8b4-43e0-ba44-b79bbf438b36/compiler_workdir/ScaledQKComparisonModule/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:138b427bf7769ac62d72b0c008b58065b93c16f2f09976be109e29ef90de7854
|
| 3 |
+
size 4940
|
neuronxcc-2.21.33363.0+82129205/MODULE_1b43d7c01692b5e7842e+9e8e849c/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11b43771b36045ed2db972b64bde026addb1f81471b9c67074c847f79c215377
|
| 3 |
+
size 390144
|
neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_b07b8b12-e364-4386-91f7-af60e9838452/compiler_workdir/OldAttentionBlock/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a16e85e684bbec7a5bb9e9df72afa8ab2fd5510493638ea9f1eea4fc86e8e50e
|
| 3 |
+
size 7822
|
neuronxcc-2.21.33363.0+82129205/MODULE_6e5f16275b9ce75f3593+27d5548e/model.neff
ADDED
|
Binary file (72.7 kB). View file
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_3fbbc5fd-f234-4f22-8d0c-aed266c1dc85/compiler_workdir/NewAttentionBlock/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a83e3ab17ecb6a05e820607c191bf82d420bac993c8802f5850cf1a9da80866
|
| 3 |
+
size 8740
|
neuronxcc-2.21.33363.0+82129205/MODULE_839f7ce0ce17e2a97166+39f22460/model.neff
ADDED
|
Binary file (82.9 kB). View file
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_ded0ff9e-5aea-499c-8629-45101ca260ae/compiler_workdir/NewSoftmaxModule/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87458fd13251d76e9632c5810e69ddd78e456d1f89d0994dba8f85036bef13e1
|
| 3 |
+
size 5596
|
neuronxcc-2.21.33363.0+82129205/MODULE_cb89e83a0ee1de5cd569+9390da5f/model.neff
ADDED
|
Binary file (42 kB). View file
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_62959f21-6073-4fd5-b612-d423dccc7c42/compiler_workdir/NewAttentionBlock/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a83e3ab17ecb6a05e820607c191bf82d420bac993c8802f5850cf1a9da80866
|
| 3 |
+
size 8740
|
neuronxcc-2.21.33363.0+82129205/MODULE_d13b2023e03adb230124+67d388cf/model.neff
ADDED
|
Binary file (82.9 kB). View file
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_b756b52a-0ddc-4f1f-b58c-ac9222676e72/compiler_workdir/OldAttentionBlock/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a16e85e684bbec7a5bb9e9df72afa8ab2fd5510493638ea9f1eea4fc86e8e50e
|
| 3 |
+
size 7822
|
neuronxcc-2.21.33363.0+82129205/MODULE_e32e6af50f0251d3e9e8+9c556093/model.neff
ADDED
|
Binary file (72.7 kB). View file
|
|
|