Synchronizing local compiler cache.
Browse files- .gitattributes +2 -0
- neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/granite/ibm-granite/granite-3.1-2b-instruct/e8aeb35344c4c108f038.json +41 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.neff +3 -0
.gitattributes
CHANGED
@@ -7681,3 +7681,5 @@ neuronxcc-2.17.194.0+d312836f/MODULE_e1cf52b9b17f1db69ab2+613edded/model.neff fi
|
|
7681 |
neuronxcc-2.17.194.0+d312836f/MODULE_f45f2eb88fadd32ecbf8+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7682 |
neuronxcc-2.17.194.0+d312836f/MODULE_553f1e4f2f3c8a659eb2+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7683 |
neuronxcc-2.17.194.0+d312836f/MODULE_97cbcc47f9c08a6a6ba1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
7681 |
neuronxcc-2.17.194.0+d312836f/MODULE_f45f2eb88fadd32ecbf8+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7682 |
neuronxcc-2.17.194.0+d312836f/MODULE_553f1e4f2f3c8a659eb2+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7683 |
neuronxcc-2.17.194.0+d312836f/MODULE_97cbcc47f9c08a6a6ba1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7684 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7685 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/granite/ibm-granite/granite-3.1-2b-instruct/e8aeb35344c4c108f038.json
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"auto_cast_type": "bf16",
|
22 |
+
"batch_size": 32,
|
23 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
24 |
+
"checkpoint_revision": "374ef54e020a3ce208c65e96d6213922a87d8952",
|
25 |
+
"compiler_type": "neuronx-cc",
|
26 |
+
"compiler_version": "2.17.194.0+d312836f",
|
27 |
+
"num_cores": 8,
|
28 |
+
"sequence_length": 4096,
|
29 |
+
"task": "text-generation"
|
30 |
+
},
|
31 |
+
"num_attention_heads": 32,
|
32 |
+
"num_hidden_layers": 40,
|
33 |
+
"num_key_value_heads": 8,
|
34 |
+
"residual_multiplier": 0.22,
|
35 |
+
"rms_norm_eps": 1e-05,
|
36 |
+
"rope_scaling": null,
|
37 |
+
"rope_theta": 5000000.0,
|
38 |
+
"tie_word_embeddings": true,
|
39 |
+
"use_cache": true,
|
40 |
+
"vocab_size": 49155
|
41 |
+
}
|
neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]
|
neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0258d9bfd3470e1043be2dcda17766548c35160d38f615820c9a9895a19ed43
|
3 |
+
size 465594
|
neuronxcc-2.17.194.0+d312836f/MODULE_3043604d5429432a4ae1+613edded/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:939cf2315fbc8c1f38b1613698cc3d3be1adc57b36b39bd94702c4a20ca24bf9
|
3 |
+
size 7793664
|
neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]
|
neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:03e63a72d2cb66d7968d51b36dc2711bef6b05509259a7e203ea20165d0dc5c0
|
3 |
+
size 594343
|
neuronxcc-2.17.194.0+d312836f/MODULE_3cb3b1b7581994ed5751+613edded/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5b7cb70578cc42bad1ea383c638250185b8d71dfe751bd723bc7a033bdbc953
|
3 |
+
size 20962304
|