Synchronizing local compiler cache.
Browse files- neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/granite/ibm-granite/granite-3.1-2b-instruct/84d1fa7d75cd5766300f.json +58 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_4be33b117e68e2de3326+a9d440f5/model.neff +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_4be33b117e68e2de3326+a9d440f5/wrapped_neff.hlo +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_50d30ce495feb34fe5c1+a9d440f5/model.neff +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_50d30ce495feb34fe5c1+a9d440f5/wrapped_neff.hlo +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_73b50e8e61409e1f8fe1+ed72d204/model.hlo_module.pb +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_73b50e8e61409e1f8fe1+ed72d204/model.neff +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_cae14ab78f54c8b4e73f+ed72d204/model.hlo_module.pb +1 -1
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_cae14ab78f54c8b4e73f+ed72d204/model.neff +1 -1
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/granite/ibm-granite/granite-3.1-2b-instruct/84d1fa7d75cd5766300f.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 4,
|
| 23 |
+
"capacity_factor": null,
|
| 24 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
| 25 |
+
"checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
|
| 26 |
+
"continuous_batching": true,
|
| 27 |
+
"enable_bucketing": false,
|
| 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": 4096,
|
| 34 |
+
"max_topk": 256,
|
| 35 |
+
"n_active_tokens": 4096,
|
| 36 |
+
"neuronxcc_version": "2.21.18209.0+043b1bf7",
|
| 37 |
+
"on_device_sampling": true,
|
| 38 |
+
"optimum_neuron_version": "0.4.1.dev0",
|
| 39 |
+
"output_logits": false,
|
| 40 |
+
"pp_degree": 1,
|
| 41 |
+
"sequence_length": 4096,
|
| 42 |
+
"speculation_length": 0,
|
| 43 |
+
"start_rank_id": 0,
|
| 44 |
+
"target": "trn1",
|
| 45 |
+
"torch_dtype": "bfloat16",
|
| 46 |
+
"tp_degree": 2
|
| 47 |
+
},
|
| 48 |
+
"num_attention_heads": 32,
|
| 49 |
+
"num_hidden_layers": 40,
|
| 50 |
+
"num_key_value_heads": 8,
|
| 51 |
+
"residual_multiplier": 0.22,
|
| 52 |
+
"rms_norm_eps": 1e-05,
|
| 53 |
+
"rope_scaling": null,
|
| 54 |
+
"rope_theta": 5000000.0,
|
| 55 |
+
"tie_word_embeddings": true,
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 49155
|
| 58 |
+
}
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4be33b117e68e2de3326+a9d440f5/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3093504
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d70d650de28d7b13defbb5f9d78b06f99ec63ba26cda803845b298cbca3d0581
|
| 3 |
size 3093504
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4be33b117e68e2de3326+a9d440f5/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3167748
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d0d1b1e40a092a3eec560354aeb1191196e8b39c2312249a4d6f358c7ef2a51
|
| 3 |
size 3167748
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_50d30ce495feb34fe5c1+a9d440f5/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1926144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af93b3d02b40084bdec001a9496f6b98115e3bfebad2f126075400a4c60fbd87
|
| 3 |
size 1926144
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_50d30ce495feb34fe5c1+a9d440f5/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2082478
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6951f83519dce668f7723834939eacc71fdb113475631340f061d1db9cd51852
|
| 3 |
size 2082478
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_73b50e8e61409e1f8fe1+ed72d204/model.hlo_module.pb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 509252
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:441fac3e96e30c894231d7f6dd20d4e89e6b4cf1295ed681a96fb69af4ea7582
|
| 3 |
size 509252
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_73b50e8e61409e1f8fe1+ed72d204/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 41585664
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccfc71e4eaf13a28e93ac88d5f63f2d222b919a9db2e9f4a966422a8f787f591
|
| 3 |
size 41585664
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_cae14ab78f54c8b4e73f+ed72d204/model.hlo_module.pb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 739366
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8511b22bc58244c0086580ec7d72532553a8f93a374268bdb1fbb737acfc3a4
|
| 3 |
size 739366
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_cae14ab78f54c8b4e73f+ed72d204/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 26133504
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ffb3d7bc9903f8346667ae3bc452aa306ed217fe3fc73cd25365d476c98eac8
|
| 3 |
size 26133504
|