Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/527899cc3b15a98022089573f6e42212494e0dbda65770e25bddf8fe48c6bd4a/e93056eae21204286577.json +59 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_4ff804c2533da46e613e+24129607/model.hlo_module.pb +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_4ff804c2533da46e613e+24129607/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/model.hlo_module.pb +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/model.neff +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/wrapped_neff.hlo +1 -1
- neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.neff +3 -0
.gitattributes
CHANGED
|
@@ -7771,3 +7771,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_f1b8afb5fe5b5d394c2b+a02c3a36/wrapped_nef
|
|
| 7771 |
neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7772 |
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7773 |
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 7771 |
neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7772 |
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7773 |
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7774 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7775 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7776 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/527899cc3b15a98022089573f6e42212494e0dbda65770e25bddf8fe48c6bd4a/e93056eae21204286577.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "ibm-granite/granite-3.1-2b-base",
|
| 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-base",
|
| 26 |
+
"checkpoint_revision": "bcf4e33e7debde263041db41869f9cdbeaae3f01",
|
| 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": "2.21.33363.0+82129205",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.6.dev3",
|
| 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": 49152
|
| 59 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_4ff804c2533da46e613e+24129607/model.hlo_module.pb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1050679
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:868cca689a34e99963ba4838928c3d05b3b7ff0c504770b6e3b057f912c499df
|
| 3 |
size 1050679
|
neuronxcc-2.21.33363.0+82129205/MODULE_4ff804c2533da46e613e+24129607/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14050304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7092c426f768f52ed7be2a5ae675118ff348d21dc203e5a969ad6da48a833fb0
|
| 3 |
size 14050304
|
neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/model.hlo_module.pb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1117688
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5e7fcdf9f65af1b99f353b41bfb54288571325ca47b98d54e3e989a576ff880
|
| 3 |
size 1117688
|
neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5387264
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29f3a621f39fe257c700d6bfc0b4b4ab251957d63970b10bb8ecf19501d9f36a
|
| 3 |
size 5387264
|
neuronxcc-2.21.33363.0+82129205/MODULE_5742ef8f5910f4e17377+a02c3a36/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5571457
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61c0c9873fdc5c05d11f5911176dcef5b6049e8a5c619cd4e51ed757c641e98e
|
| 3 |
size 5571457
|
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f42c09e1757c1b8cc6c9915888f6eec10efda443cff236c078572ce31ea01c7a
|
| 3 |
+
size 1072136
|
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c068f9b76116c5732ae7ccb9b27d847d79efe553f050eb104d6cbdce681acbe4
|
| 3 |
+
size 5141504
|
neuronxcc-2.21.33363.0+82129205/MODULE_6781be5fc2d8529ad204+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccfd0debc2b3578ea74099a763ad76bf4c23a85b79dc90786556ff595364d0b3
|
| 3 |
+
size 5326023
|
neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/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/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d3aeb00acbd6df6d692c58e05895f2844334b18359ca902b82a8e12bf74f210
|
| 3 |
+
size 1405263
|
neuronxcc-2.21.33363.0+82129205/MODULE_f100cf6bbe2c168dffc8+6170d8e1/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52809966c77105958fdb72bfc5dc0a79bf3c729153426cb0e4aadffa361f9883
|
| 3 |
+
size 111524864
|