Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/granite/ibm-granite/granite-3.1-2b-instruct/c3e37d9fcb69595ff392.json +59 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/compile_flags.json +1 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.done +0 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.hlo_module.pb +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.neff +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/compile_flags.json +1 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.done +0 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.hlo_module.pb +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.neff +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes
CHANGED
|
@@ -11233,3 +11233,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c1d6512f90c8fdc2230c+a9d440f5/wrapped_nef
|
|
| 11233 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_313f0d96d7fbbab7bf88+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11234 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b95c5048fdc77759bd84+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11235 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b95c5048fdc77759bd84+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 11233 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_313f0d96d7fbbab7bf88+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11234 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b95c5048fdc77759bd84+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11235 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b95c5048fdc77759bd84+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 11236 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11237 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 11238 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/granite/ibm-granite/granite-3.1-2b-instruct/c3e37d9fcb69595ff392.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 |
+
"embedding_multiplier": 12.0,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"logits_scaling": 8.0,
|
| 17 |
+
"max_position_embeddings": 131072,
|
| 18 |
+
"mlp_bias": false,
|
| 19 |
+
"model_type": "granite",
|
| 20 |
+
"neuron": {
|
| 21 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 22 |
+
"batch_size": 1,
|
| 23 |
+
"capacity_factor": null,
|
| 24 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
| 25 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
| 26 |
+
"continuous_batching": false,
|
| 27 |
+
"enable_bucketing": false,
|
| 28 |
+
"ep_degree": 1,
|
| 29 |
+
"fused_qkv": true,
|
| 30 |
+
"glu_mlp": true,
|
| 31 |
+
"local_ranks_size": 8,
|
| 32 |
+
"logical_nc_config": 1,
|
| 33 |
+
"max_batch_size": 1,
|
| 34 |
+
"max_context_length": 4096,
|
| 35 |
+
"max_topk": 256,
|
| 36 |
+
"n_active_tokens": 4096,
|
| 37 |
+
"neuronxcc_version": "2.21.18209.0+043b1bf7",
|
| 38 |
+
"on_device_sampling": true,
|
| 39 |
+
"optimum_neuron_version": "0.3.1.dev7",
|
| 40 |
+
"output_logits": false,
|
| 41 |
+
"pp_degree": 1,
|
| 42 |
+
"sequence_length": 4096,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": null,
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"tp_degree": 8
|
| 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-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1d9b3a9753cd52ab90b20d91b3f9f82ebec3623b48e46d556ccedc2ca279fa6
|
| 3 |
+
size 908283
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_611f97b61453e4e426db+ed72d204/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:260098c80e2d12472e21aaf255c07a3efcbcbf177060c9fbd1fa6fc8705e4cc0
|
| 3 |
+
size 25078784
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd4c9c1909e9c4adfba882d82640bf7b19f7c0f00a7bbff19cec52b4291dfb51
|
| 3 |
+
size 1013581
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a62bf8a0b2151458abdbb9116ffe2863878df6c7e3fd67c755c72c56c56cc3bb
|
| 3 |
+
size 1772544
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_de6f676f49e84c0b5be2+a9d440f5/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7f23a19b6067c7070f7949437f4d5e4c3b74b2d94055ee223033f3cb3c4ff00
|
| 3 |
+
size 1957303
|