Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +21 -0
- neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/3722d0e82203fbbe93fe.json +73 -0
- neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/a8be13be525f2d91669b.json +73 -0
- neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/ca40c0099b06c7de4aa6.json +73 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/wrapped_neff.hlo +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.neff +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.neff +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_804dc8509b11b248fd01+bfe5714b/compile_flags.json +1 -0
.gitattributes
CHANGED
@@ -3181,3 +3181,24 @@ neuronxcc-2.17.194.0+d312836f/MODULE_5a10198534c5f2725fd7+165e9558/wrapped_neff.
|
|
3181 |
neuronxcc-2.17.194.0+d312836f/MODULE_bc0dc6318052d18d4f59+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3182 |
neuronxcc-2.17.194.0+d312836f/MODULE_bc0dc6318052d18d4f59+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3183 |
neuronxcc-2.17.194.0+d312836f/MODULE_b73c22227b6353be7e03+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3181 |
neuronxcc-2.17.194.0+d312836f/MODULE_bc0dc6318052d18d4f59+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3182 |
neuronxcc-2.17.194.0+d312836f/MODULE_bc0dc6318052d18d4f59+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3183 |
neuronxcc-2.17.194.0+d312836f/MODULE_b73c22227b6353be7e03+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3184 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3185 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3186 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3187 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3188 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3189 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.neff filter=lfs diff=lfs merge=lfs -text
|
3190 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3191 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3192 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3193 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3194 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_804dc8509b11b248fd01+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3195 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_9840ca94af4106910064+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3196 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_a24e3ed896dae389d4f2+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3197 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_b5cb392eb50260fa24a3+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3198 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_ba9e5f7f86364cb7b38d+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3199 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_ba9e5f7f86364cb7b38d+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3200 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_c36623907d50079ba312+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
|
3201 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_cb76f3926e2853557294+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
|
3202 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_cb76f3926e2853557294+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
3203 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_dea3fa0fa1232db56e94+26ac6be0/model.neff filter=lfs diff=lfs merge=lfs -text
|
3204 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_f3e8207126f92d912816+26ac6be0/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/3722d0e82203fbbe93fe.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"GraniteForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"attention_multiplier": 1.0,
|
11 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 1,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": false,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 1,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.17.194.0+d312836f",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.0.dev2",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "float16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "float16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/a8be13be525f2d91669b.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"GraniteForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"attention_multiplier": 1.0,
|
11 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 1,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": false,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 1,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.17.194.0+d312836f",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": true,
|
47 |
+
"optimum_neuron_version": "0.3.0.dev2",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "bfloat16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "bfloat16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/ca40c0099b06c7de4aa6.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"GraniteForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"attention_multiplier": 1.0,
|
11 |
+
"embedding_multiplier": 1.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 32,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 64,
|
16 |
+
"logits_scaling": 1.0,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"_serialized_key": "NxDNeuronConfig",
|
22 |
+
"async_mode": false,
|
23 |
+
"attn_kernel_enabled": false,
|
24 |
+
"batch_size": 2,
|
25 |
+
"capacity_factor": null,
|
26 |
+
"cc_pipeline_tiling_factor": 2,
|
27 |
+
"checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
|
28 |
+
"checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
|
29 |
+
"continuous_batching": true,
|
30 |
+
"enable_bucketing": false,
|
31 |
+
"ep_degree": 1,
|
32 |
+
"flash_decoding_enabled": false,
|
33 |
+
"fused_qkv": true,
|
34 |
+
"glu_mlp": true,
|
35 |
+
"is_chunked_prefill": false,
|
36 |
+
"local_ranks_size": 2,
|
37 |
+
"logical_nc_config": 1,
|
38 |
+
"max_batch_size": 2,
|
39 |
+
"max_context_length": 100,
|
40 |
+
"max_topk": 256,
|
41 |
+
"mlp_kernel_enabled": false,
|
42 |
+
"mlp_kernel_fuse_residual_add": false,
|
43 |
+
"n_active_tokens": 100,
|
44 |
+
"neuronxcc_version": "2.17.194.0+d312836f",
|
45 |
+
"num_cores_per_group": 1,
|
46 |
+
"on_device_sampling": false,
|
47 |
+
"optimum_neuron_version": "0.3.0.dev2",
|
48 |
+
"output_logits": false,
|
49 |
+
"padding_side": "right",
|
50 |
+
"pp_degree": 1,
|
51 |
+
"qk_layernorm": false,
|
52 |
+
"qkv_kernel_enabled": false,
|
53 |
+
"rpl_reduce_dtype": "float16",
|
54 |
+
"sequence_length": 100,
|
55 |
+
"sequence_parallel_enabled": false,
|
56 |
+
"speculation_length": 0,
|
57 |
+
"start_rank_id": 0,
|
58 |
+
"target": null,
|
59 |
+
"torch_dtype": "float16",
|
60 |
+
"tp_degree": 2,
|
61 |
+
"vocab_parallel": false
|
62 |
+
},
|
63 |
+
"num_attention_heads": 4,
|
64 |
+
"num_hidden_layers": 2,
|
65 |
+
"num_key_value_heads": 4,
|
66 |
+
"residual_multiplier": 1.0,
|
67 |
+
"rms_norm_eps": 1e-06,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 10000.0,
|
70 |
+
"tie_word_embeddings": false,
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 49152
|
73 |
+
}
|
neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9d37a05e9076514bd247ba643d2dfb34bdca5f1924f390835ca6431fd5d60cf9
|
3 |
+
size 80261
|
neuronxcc-2.17.194.0+d312836f/MODULE_014569533dcbafbc3ea9+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5670fe5126454249b42797df432aedae725b5486cb2d590368215780fceef68
|
3 |
+
size 205824
|
neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b1379bf9d9198be3627ce2a98a8bc77592586561dba15736ada02f1a62b03546
|
3 |
+
size 46622
|
neuronxcc-2.17.194.0+d312836f/MODULE_02f045f6902463c49bce+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4bf046ebf5a86a78e87113f066962dd06a177293863e91ecea5b65641c65b57d
|
3 |
+
size 144384
|
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
|
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2e1d6b91a43d0f9f90c646826c509fe753b4689880129a8155140db6a1621186
|
3 |
+
size 46541
|
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4868ab12ff273c74e0a4eb1a1631f62a3ff9692ce6df293df4a6d0d808573392
|
3 |
+
size 144384
|
neuronxcc-2.17.194.0+d312836f/MODULE_0bbe60dde8eaacbc8218+165e9558/wrapped_neff.hlo
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b8905529633425558763cbf12d6c6817d71ab06370472e5e67169aa7a59119f8
|
3 |
+
size 152045
|
neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0991e707bfd4b423cddb443d20b4f5aec4b1c262e379e65b64c27b13445ce083
|
3 |
+
size 7099
|
neuronxcc-2.17.194.0+d312836f/MODULE_0f88705903403514996d+431f5505/model.neff
ADDED
Binary file (42 kB). View file
|
|
neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09596247b06e97404db0e3f45a6e58f5ebefe9afc54db0909b7232a603298766
|
3 |
+
size 53803
|
neuronxcc-2.17.194.0+d312836f/MODULE_1215feca19e3858f9ef6+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a688781c302c2ccfe6ce93506b7149064bbc731423ba74ef2621e257d3158b7
|
3 |
+
size 164864
|
neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e79cc303169501720491636df85a8ef222e195d022e9b0af0c8c9a23aa555c2
|
3 |
+
size 81609
|
neuronxcc-2.17.194.0+d312836f/MODULE_1e78bd200a100a9daaf0+26ac6be0/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f9f58954fc76cbb93a6552b432728a6fd38d3787465f175ec0145193a4a866dc
|
3 |
+
size 359424
|
neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2df65817e2703ab6983c175431e26c37a0ab5ee333592a6575d221e97a643fc5
|
3 |
+
size 81115
|
neuronxcc-2.17.194.0+d312836f/MODULE_26b84de33e6524d9507a+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:629254561002f6a231db1c9da209fc813092957c8f2d614063721685a2bb1b47
|
3 |
+
size 226304
|
neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea173b3def731778aae171bcbbb909ddc1467d1eb8304f5dde6f6ef914e45ec0
|
3 |
+
size 87577
|
neuronxcc-2.17.194.0+d312836f/MODULE_41612d58bf9ba6f268b0+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:747f2f2298b6ad897eedb7a9ae8b4184a1dad89956bfffa354da44231c8d1cd9
|
3 |
+
size 246784
|
neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cecb8b2e6d9a2ea5d80a7f910d95611058e25448a5765cc2bdf5e2d5a6465519
|
3 |
+
size 7092
|
neuronxcc-2.17.194.0+d312836f/MODULE_61794b8717d8b5a8853d+431f5505/model.neff
ADDED
Binary file (42 kB). View file
|
|
neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:97758a2b3a85873d3978bdcc4ad957312312aadec91e4e6ed596de7aa41731fc
|
3 |
+
size 47478
|
neuronxcc-2.17.194.0+d312836f/MODULE_7af318ed51d57f96cca6+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea7afc2cf7419cc0ec0088f52f3ca452408f18ce12d91004649ec710e00ed064
|
3 |
+
size 154624
|
neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|
neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:01367b6b5a410c495ff9d3fa49e16ba09bb17b22715607ce41e30dcaf108db9b
|
3 |
+
size 87577
|
neuronxcc-2.17.194.0+d312836f/MODULE_7d17be30edb500bd0e79+bfe5714b/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:55614536bfcab0b8aa8783521e0ebffff16849c91730f0a25f0137121fa6b906
|
3 |
+
size 246784
|
neuronxcc-2.17.194.0+d312836f/MODULE_804dc8509b11b248fd01+bfe5714b/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
|