Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/mixtral/mistralai/Mixtral-8x22B-Instruct-v0.1/e19b3890d1d3933e8b87.json +58 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/model.done +0 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/model.neff +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/compile_flags.json +1 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.done +0 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.hlo_module.pb +3 -0
- neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.neff +3 -0
.gitattributes
CHANGED
|
@@ -12217,3 +12217,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+ca355898/model.neff
|
|
| 12217 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 12218 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+ca355898/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 12219 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 12217 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 12218 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+ca355898/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 12219 |
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+ca355898/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 12220 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 12221 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 12222 |
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/mixtral/mistralai/Mixtral-8x22B-Instruct-v0.1/e19b3890d1d3933e8b87.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "mistralai/Mixtral-8x22B-Instruct-v0.1",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"MixtralForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 6144,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 16384,
|
| 14 |
+
"max_position_embeddings": 65536,
|
| 15 |
+
"model_type": "mixtral",
|
| 16 |
+
"neuron": {
|
| 17 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 18 |
+
"batch_size": 1,
|
| 19 |
+
"capacity_factor": null,
|
| 20 |
+
"checkpoint_id": "mistralai/Mixtral-8x22B-Instruct-v0.1",
|
| 21 |
+
"checkpoint_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
|
| 22 |
+
"continuous_batching": false,
|
| 23 |
+
"enable_bucketing": false,
|
| 24 |
+
"ep_degree": 1,
|
| 25 |
+
"fused_qkv": false,
|
| 26 |
+
"glu_mlp": true,
|
| 27 |
+
"local_ranks_size": 64,
|
| 28 |
+
"max_batch_size": 1,
|
| 29 |
+
"max_context_length": 4096,
|
| 30 |
+
"max_topk": 256,
|
| 31 |
+
"n_active_tokens": 4096,
|
| 32 |
+
"neuronxcc_version": "2.21.18209.0+043b1bf7",
|
| 33 |
+
"on_device_sampling": false,
|
| 34 |
+
"optimum_neuron_version": "0.4.1.dev0",
|
| 35 |
+
"output_logits": false,
|
| 36 |
+
"pp_degree": 1,
|
| 37 |
+
"sequence_length": 4096,
|
| 38 |
+
"speculation_length": 0,
|
| 39 |
+
"start_rank_id": 0,
|
| 40 |
+
"target": "trn2",
|
| 41 |
+
"torch_dtype": "bfloat16",
|
| 42 |
+
"tp_degree": 64
|
| 43 |
+
},
|
| 44 |
+
"num_attention_heads": 48,
|
| 45 |
+
"num_experts_per_tok": 2,
|
| 46 |
+
"num_hidden_layers": 56,
|
| 47 |
+
"num_key_value_heads": 8,
|
| 48 |
+
"num_local_experts": 8,
|
| 49 |
+
"output_router_logits": false,
|
| 50 |
+
"rms_norm_eps": 1e-05,
|
| 51 |
+
"rope_theta": 1000000.0,
|
| 52 |
+
"router_aux_loss_coef": 0.001,
|
| 53 |
+
"router_jitter_noise": 0.0,
|
| 54 |
+
"sliding_window": null,
|
| 55 |
+
"tie_word_embeddings": false,
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 32768
|
| 58 |
+
}
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4d3419e94455047e02d575ddf8e68bf021e83e214e7dd4ccd1929d7d34cae46
|
| 3 |
+
size 4824064
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_887a24107396c649d6c0+0b01cb42/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9ac5bc7cc02db8c1da1d9c3ded18651d1bcc8b6e7a95e52c483f3f211dc1087
|
| 3 |
+
size 5125878
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e37c00ff1c4a167c5b81da7ae9af1041c7048efa5d4e29af8a074ed12264642
|
| 3 |
+
size 2629000
|
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ed84bba0a6ce657ea8a5+be13b572/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5bc55bff40597422a3e64f51de70059bfb328a6b8fcb9fffd375c518c76d94c
|
| 3 |
+
size 3625984
|