Upload folder using huggingface_hub

-_0_0_0/after_recompile_pre_grad_3.txt

@@ -0,0 +1,13 @@
+class GraphModule(torch.nn.Module):
+    def forward(self, L_w_: "f32[4096, 8192][8192, 1]cuda:0", L_scale_: "f32[4096][1]cuda:0", L_zero_point_: "i32[4096][1]cuda:0"):
+        l_w_ = L_w_
+        l_scale_ = L_scale_
+        l_zero_point_ = L_zero_point_
+        
+         # File: /shared_volume/repos/quark/bench_qdq.py:8 in run_scaled_fake_quantize, code: return scaled_fake_quantize(quant_dtype, w, scale, zero_point, axis, group_size, quant_min, quant_max, round_mode, qscheme, mx_element_dtype)
+        function_ctx = torch.autograd.function.FunctionCtx();  function_ctx = None
+        
+         # File: /shared_volume/repos/quark/quark/torch/kernel/__init__.py:168 in forward, code: return ops.quark.scaled_fake_quantize(quant_dtype, inputs, scale, zero_point, axis, group_size, quant_min,
+        scaled_fake_quantize: "f32[4096, 8192][8192, 1]cuda:0" = torch.ops.quark.scaled_fake_quantize('int4', l_w_, l_scale_, l_zero_point_, 0, 0, 0, 15, 0, 'per_channel', 'haha');  l_w_ = l_scale_ = l_zero_point_ = None
+        return (scaled_fake_quantize,)
+        

-_0_0_0/aot_forward_graph_fw_metadata_4.txt

@@ -0,0 +1,57 @@
+ViewAndMutationMeta(input_info=[InputAliasInfo(is_leaf=True,
+                                              mutates_data=False,
+                                              mutates_metadata=False,
+                                              mutations_hidden_from_autograd=True,
+                                              mutations_under_no_grad_or_inference_mode=False,
+                                              mutation_inductor_storage_resize=False,
+                                              mutates_storage_metadata=False,
+                                              requires_grad=False,
+                                              keep_input_mutations=True),
+                               InputAliasInfo(is_leaf=True,
+                                              mutates_data=False,
+                                              mutates_metadata=False,
+                                              mutations_hidden_from_autograd=True,
+                                              mutations_under_no_grad_or_inference_mode=False,
+                                              mutation_inductor_storage_resize=False,
+                                              mutates_storage_metadata=False,
+                                              requires_grad=False,
+                                              keep_input_mutations=True),
+                               InputAliasInfo(is_leaf=True,
+                                              mutates_data=False,
+                                              mutates_metadata=False,
+                                              mutations_hidden_from_autograd=True,
+                                              mutations_under_no_grad_or_inference_mode=False,
+                                              mutation_inductor_storage_resize=False,
+                                              mutates_storage_metadata=False,
+                                              requires_grad=False,
+                                              keep_input_mutations=True)],
+                    output_info=[OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
+                                                raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
+                                                base_idx=None,
+                                                dynamic_dims=set(),
+                                                requires_grad=False,
+                                                functional_tensor=None)],
+                    num_intermediate_bases=0,
+                    keep_input_mutations=True,
+                    traced_tangents=[],
+                    subclass_inp_meta=[PlainTensorMeta(unwrapped_idx=0,
+                                                      memory_format=None),
+                                      PlainTensorMeta(unwrapped_idx=1,
+                                                      memory_format=None),
+                                      PlainTensorMeta(unwrapped_idx=2,
+                                                      memory_format=None)],
+                    subclass_fw_graph_out_meta=[PlainTensorMeta(unwrapped_idx=0,
+                                                               memory_format=None)],
+                    subclass_tangent_meta=[],
+                    is_train=False,
+                    traced_tangent_metas=None,
+                    num_symints_saved_for_bw=None,
+                    grad_enabled_mutation=None,
+                    deterministic=None,
+                    static_input_indices=[],
+                    tokens={},
+                    indices_of_inputs_that_requires_grad_with_mutations_in_bw=[],
+                    bw_donated_idxs=None,
+                    num_backward_tokens=0,
+                    num_graphsafe_rng_states=0,
+                    graphsafe_rng_state_index=None)

-_0_0_0/aot_inference_graph_5.txt

@@ -0,0 +1,6 @@
+class <lambda>(torch.nn.Module):
+    def forward(self, arg0_1: "f32[4096, 8192][8192, 1]cuda:0", arg1_1: "f32[4096][1]cuda:0", arg2_1: "i32[4096][1]cuda:0"):
+         # File: /shared_volume/repos/quark/quark/torch/kernel/__init__.py:168 in forward, code: return ops.quark.scaled_fake_quantize(quant_dtype, inputs, scale, zero_point, axis, group_size, quant_min,
+        scaled_fake_quantize: "f32[4096, 8192][8192, 1]cuda:0" = torch.ops.quark.scaled_fake_quantize.default('int4', arg0_1, arg1_1, arg2_1, 0, 0, 0.0, 15.0, 0, 'per_channel', 'haha');  arg0_1 = arg1_1 = arg2_1 = None
+        return (scaled_fake_quantize,)
+        

-_0_0_0/aotautograd_cache_bypass_9.json

@@ -0,0 +1,8 @@
+{
+  "cache_bypass_hard_exception": false,
+  "cache_bypass_reason": "Unsupported call_function target <class 'torch.autograd.function.FunctionCtx'>. \n Function module: torch.autograd.function, \nFunction name: FunctionCtx",
+  "cache_state": "bypass",
+  "compile_id": "0/0",
+  "components": [],
+  "key": null
+}

-_0_0_0/before_recompile_pre_grad_2.txt

@@ -0,0 +1,13 @@
+class GraphModule(torch.nn.Module):
+    def forward(self, L_w_: "f32[4096, 8192][8192, 1]cuda:0", L_scale_: "f32[4096][1]cuda:0", L_zero_point_: "i32[4096][1]cuda:0"):
+        l_w_ = L_w_
+        l_scale_ = L_scale_
+        l_zero_point_ = L_zero_point_
+        
+         # File: /shared_volume/repos/quark/bench_qdq.py:8 in run_scaled_fake_quantize, code: return scaled_fake_quantize(quant_dtype, w, scale, zero_point, axis, group_size, quant_min, quant_max, round_mode, qscheme, mx_element_dtype)
+        function_ctx = torch.autograd.function.FunctionCtx();  function_ctx = None
+        
+         # File: /shared_volume/repos/quark/quark/torch/kernel/__init__.py:168 in forward, code: return ops.quark.scaled_fake_quantize(quant_dtype, inputs, scale, zero_point, axis, group_size, quant_min,
+        scaled_fake_quantize: "f32[4096, 8192][8192, 1]cuda:0" = torch.ops.quark.scaled_fake_quantize('int4', l_w_, l_scale_, l_zero_point_, 0, 0, 0, 15, 0, 'per_channel', 'haha');  l_w_ = l_scale_ = l_zero_point_ = None
+        return (scaled_fake_quantize,)
+        

-_0_0_0/compilation_metrics_11.html

@@ -0,0 +1,152 @@
+
+<html>
+<head>
+    <style>
+    
+table td { vertical-align: top; }
+
+.stack-trie { white-space: nowrap; font-family: monospace; }
+.stack-trie ul { padding-left: 1ch;  }
+.stack-trie li { margin-left: 1ch; list-style-type: none;  }
+.stack-trie .marker {
+  cursor: pointer;
+}
+.stack-trie .marker.collapsed::before {
+  content: &quot;+ &quot;;
+}
+.stack-trie .marker:not(.collapsed)::before {
+  content: &quot;- &quot;;
+}
+.stack-trie a { text-decoration: none; }
+.stack-trie a:hover { text-decoration: underline; }
+.status-missing { background-color: purple; color: white; }
+.status-error { background-color: red; color: white; }
+.status-empty { background-color: white; color: black; }
+.status-ok { background-color: green; color: white; }
+.status-break { background-color: lime; color: black; }
+summary::-webkit-details-marker { color: #00ACF3; font-size: 125%; margin-right: 2px; }
+summary:focus { outline-style: none; }
+article &gt; details &gt; summary { font-size: 28px; margin-top: 16px; }
+details &gt; p { margin-left: 24px; }
+details details summary { font-size: 16px; }
+
+    </style>
+    <title>Compilation Metrics</title>
+    <base href="..">
+</head>
+<body>
+    <h1>Compilation Info for [0/0] </h1>
+    <p><details><summary>Stack</summary><div class='stack-trie'><ul><li>/shared_volume/repos/quark/bench_qdq.py:7 in run_scaled_fake_quantize<br>&nbsp;&nbsp;&nbsp;&nbsp;</li>
+</ul></div></details></p>
+    <h2>Output files:</h2>
+    <ul>
+        
+            <li><a href="-_0_0_0/dynamo_output_graph_0.txt">dynamo_output_graph_0.txt</a> (0)</li>
+        
+            <li><a href="-_0_0_0/inductor_pre_grad_graph_1.txt">inductor_pre_grad_graph_1.txt</a> (1)</li>
+        
+            <li><a href="-_0_0_0/before_recompile_pre_grad_2.txt">before_recompile_pre_grad_2.txt</a> (2)</li>
+        
+            <li><a href="-_0_0_0/after_recompile_pre_grad_3.txt">after_recompile_pre_grad_3.txt</a> (3)</li>
+        
+            <li><a href="-_0_0_0/aot_forward_graph_fw_metadata_4.txt">aot_forward_graph_fw_metadata_4.txt</a> (4)</li>
+        
+            <li><a href="-_0_0_0/aot_inference_graph_5.txt">aot_inference_graph_5.txt</a> (5)</li>
+        
+            <li><a href="-_0_0_0/torch._functorch.config_6.txt">torch._functorch.config_6.txt</a> (6)</li>
+        
+            <li><a href="-_0_0_0/inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html">inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html</a> (7)</li>
+        
+            <li><a href="-_0_0_0/fx_graph_cache_hit_8.json">fx_graph_cache_hit_8.json</a> (8)</li>
+        
+            <li><a href="-_0_0_0/aotautograd_cache_bypass_9.json">aotautograd_cache_bypass_9.json</a> (9)</li>
+        
+            <li><a href="-_0_0_0/dynamo_cpp_guards_str_10.txt">dynamo_cpp_guards_str_10.txt</a> (10)</li>
+        
+    </ul>
+    <h2>Stack</h2>
+    <details><summary>Stack</summary><div class='stack-trie'><ul><li>/shared_volume/repos/quark/bench_qdq.py:161 in &lt;module&gt;<br>&nbsp;&nbsp;&nbsp;&nbsp;mean, median = do_bench(run_scaled_fake_quantize_comp, kwargs_scaled_fake_quantize, num_runs=num_runs, num_warmup=num_warmup, name="quark qdq")</li>
+<li>/shared_volume/repos/quark/bench_qdq.py:70 in do_bench<br>&nbsp;&nbsp;&nbsp;&nbsp;f(**kwargs)</li>
+<li>/shared_volume/repos/quark/bench_qdq.py:7 in run_scaled_fake_quantize<br>&nbsp;&nbsp;&nbsp;&nbsp;</li>
+</ul></div></details>
+    <h2>Compile Time(seconds)</h2>
+    <p>Entire Frame <abbr title="Total time spent in convert_frame function">[?]</abbr>: 1.068891</div>
+    <p>Backend <abbr title="Time spent running the backend compiler">[?]</abbr>: 0.992908</div>
+    
+    <p>Inductor <abbr title="Total time spent running inductor">[?]</abbr>: 0.414598</div>
+    
+    
+    <div>Dynamo Time Before Restart <abbr title="Total time spent restarting dynamo analysis">[?]</abbr>: 0.0</div>
+    <h2>Restarts and Failures</h2>
+    
+    <p> No failures! </p>
+    
+    
+    <p> No restarts! </p>
+    
+    <h2>Cache Metrics</h2>
+    <p>Cache Size: 0</p>
+    <p>Accumulated Cache Size: 0</p>
+    <h2>Graph Metrics</h2>
+    <p>Guard Count: 21</p>
+    <p>Shape Env Guards: 0</p>
+    <p>Graph Ops: 2</p>
+    <p>Graph Nodes: 6</p>
+    <p>Graph Inputs: 3</p>
+    <h2> Custom Ops </h2>
+    
+    
+    <p> Non-Compliant Custom Ops:</p>
+    
+    <li> <code> quark::scaled_fake_quantize </code> </li>
+    
+    
+    <h2>Symbolic shape specializations</h2>
+    <table>
+    <tr>
+        <th>Sym</th> <th>Source(s)</th> <th>Value</th> <th>User stack</th> <th>Framework stack</th>
+    </tr>
+    
+    </table>
+    <h2>Guards added fast</h2>
+    <table>
+    <tr>
+        <th>Expr</th> <th>User stack</th> <th>Framework stack</th>
+    </tr>
+    
+    </table>
+    
+    <script>
+    document.addEventListener('DOMContentLoaded', function() {
+
+        // Append the current URL's query parameters to all relative links on the page
+        const queryParams = new URLSearchParams(window.location.search);
+        if (queryParams.size === 0) return url; // No query params, return original URL
+
+        function appendQueryParams(url) {
+            const newURL = new URL((new Request(url)).url);  // new URL(<relative URL>) but it actually works
+            const newSearchParams = new URLSearchParams(newURL.searchParams);
+            console.log(newURL.searchParams);
+            console.log(newSearchParams);
+
+            // Append query parameters
+            for (const [key, value] of queryParams) {
+                newSearchParams.set(key, value);
+            }
+
+            newURL.search = newSearchParams;
+            return newURL;
+        }
+
+        // Select all relative links on the page
+        const relativeLinks = document.querySelectorAll('a[href]:not([href^="http://"]):not([href^="https://"]):not([href^="\#"])');
+
+        // Append query parameters to each relative link
+        relativeLinks.forEach((link) => {
+            link.setAttribute("href", appendQueryParams(link.getAttribute("href")))
+        });
+    });
+    </script>
+
+</body>
+</html>

-_0_0_0/dynamo_cpp_guards_str_10.txt

@@ -0,0 +1,47 @@
+
+TREE_GUARD_MANAGER:
++- RootGuardManager
+| +- DEFAULT_DEVICE: utils_device.CURRENT_DEVICE == None                           # _dynamo/output_graph.py:520 in init_ambient_guards
+| +- GLOBAL_STATE: ___check_global_state()
+| +- TORCH_FUNCTION_MODE_STACK: ___check_torch_function_mode_stack()
+| +- GuardManager: source=L['w'], accessed_by=FrameLocalsGuardAccessor(key='w', framelocals_idx=1)
+| | +- TENSOR_MATCH: check_tensor(L['w'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.float32, device=0, requires_grad=False, size=[4096, 8192], stride=[8192, 1])
+| | +- NO_HASATTR: hasattr(L['w'], '_dynamo_dynamic_indices') == False         
+| | +- NO_TENSOR_ALIASING: check_no_aliasing(L['w'], L['scale'], L['zero_point'])
+| +- GuardManager: source=L['axis'], accessed_by=FrameLocalsGuardAccessor(key='axis', framelocals_idx=4)
+| | +- EQUALS_MATCH: L['axis'] == 0                                              
+| +- GuardManager: source=L['scale'], accessed_by=FrameLocalsGuardAccessor(key='scale', framelocals_idx=2)
+| | +- TENSOR_MATCH: check_tensor(L['scale'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.float32, device=0, requires_grad=False, size=[4096], stride=[1])
+| | +- NO_HASATTR: hasattr(L['scale'], '_dynamo_dynamic_indices') == False     
+| | +- NO_TENSOR_ALIASING
+| +- GuardManager: source=L['qscheme'], accessed_by=FrameLocalsGuardAccessor(key='qscheme', framelocals_idx=9)
+| | +- EQUALS_MATCH: L['qscheme'] == 'per_channel'                               
+| +- GuardManager: source=L['quant_max'], accessed_by=FrameLocalsGuardAccessor(key='quant_max', framelocals_idx=7)
+| | +- EQUALS_MATCH: L['quant_max'] == 15                                        
+| +- GuardManager: source=L['quant_min'], accessed_by=FrameLocalsGuardAccessor(key='quant_min', framelocals_idx=6)
+| | +- EQUALS_MATCH: L['quant_min'] == 0                                         
+| +- GuardManager: source=L['group_size'], accessed_by=FrameLocalsGuardAccessor(key='group_size', framelocals_idx=5)
+| | +- EQUALS_MATCH: L['group_size'] == 0                                        
+| +- GuardManager: source=L['round_mode'], accessed_by=FrameLocalsGuardAccessor(key='round_mode', framelocals_idx=8)
+| | +- EQUALS_MATCH: L['round_mode'] == 0                                        
+| +- GuardManager: source=L['zero_point'], accessed_by=FrameLocalsGuardAccessor(key='zero_point', framelocals_idx=3)
+| | +- TENSOR_MATCH: check_tensor(L['zero_point'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.int32, device=0, requires_grad=False, size=[4096], stride=[1])
+| | +- NO_HASATTR: hasattr(L['zero_point'], '_dynamo_dynamic_indices') == False
+| | +- NO_TENSOR_ALIASING
+| +- GuardManager: source=L['quant_dtype'], accessed_by=FrameLocalsGuardAccessor(key='quant_dtype', framelocals_idx=0)
+| | +- EQUALS_MATCH: L['quant_dtype'] == 'int4'                                  
+| +- GuardManager: source=L['mx_element_dtype'], accessed_by=FrameLocalsGuardAccessor(key='mx_element_dtype', framelocals_idx=10)
+| | +- EQUALS_MATCH: L['mx_element_dtype'] == 'haha'                             
+| +- GuardManager: source=G, accessed_by=GlobalsGuardAccessor
+| | +- GuardManager: source=G['scaled_fake_quantize'], accessed_by=DictGetItemGuardAccessor('scaled_fake_quantize')
+| | | +- ID_MATCH: ___check_obj_id(G['scaled_fake_quantize'], 140615015598720) 
+| | +- GuardManager: source=G['__import_quark_dot_torch_dot_kernel'], accessed_by=DictGetItemGuardAccessor('__import_quark_dot_torch_dot_kernel')
+| | | +- ID_MATCH: ___check_obj_id(G['__import_quark_dot_torch_dot_kernel'], 140649485511360)
+| | | +- GuardManager: source=G['__import_quark_dot_torch_dot_kernel'].ops, accessed_by=GetAttrGuardAccessor(ops)
+| | | | +- ID_MATCH: ___check_obj_id(G['__import_quark_dot_torch_dot_kernel'].ops, 140649515082000)
+| | | | +- GuardManager: source=G['__import_quark_dot_torch_dot_kernel'].ops.quark, accessed_by=GetAttrGuardAccessor(quark)
+| | | | | +- ID_MATCH: ___check_obj_id(G['__import_quark_dot_torch_dot_kernel'].ops.quark, 140649455882608)
+| | | | | +- GuardManager: source=G['__import_quark_dot_torch_dot_kernel'].ops.quark.scaled_fake_quantize, accessed_by=GetAttrGuardAccessor(scaled_fake_quantize)
+| | | | | | +- ID_MATCH: ___check_obj_id(G['__import_quark_dot_torch_dot_kernel'].ops.quark.scaled_fake_quantize, 140614946665776)
+
+Guard latency = 0.84 us

-_0_0_0/dynamo_output_graph_0.txt

@@ -0,0 +1,13 @@
+class GraphModule(torch.nn.Module):
+    def forward(self, L_w_: "f32[4096, 8192][8192, 1]cuda:0", L_scale_: "f32[4096][1]cuda:0", L_zero_point_: "i32[4096][1]cuda:0"):
+        l_w_ = L_w_
+        l_scale_ = L_scale_
+        l_zero_point_ = L_zero_point_
+        
+         # File: /shared_volume/repos/quark/bench_qdq.py:8 in run_scaled_fake_quantize, code: return scaled_fake_quantize(quant_dtype, w, scale, zero_point, axis, group_size, quant_min, quant_max, round_mode, qscheme, mx_element_dtype)
+        function_ctx = torch.autograd.function.FunctionCtx();  function_ctx = None
+        
+         # File: /shared_volume/repos/quark/quark/torch/kernel/__init__.py:168 in forward, code: return ops.quark.scaled_fake_quantize(quant_dtype, inputs, scale, zero_point, axis, group_size, quant_min,
+        scaled_fake_quantize: "f32[4096, 8192][8192, 1]cuda:0" = torch.ops.quark.scaled_fake_quantize('int4', l_w_, l_scale_, l_zero_point_, 0, 0, 0, 15, 0, 'per_channel', 'haha');  l_w_ = l_scale_ = l_zero_point_ = None
+        return (scaled_fake_quantize,)
+        

-_0_0_0/fx_graph_cache_hit_8.json

@@ -0,0 +1,315 @@
+{
+  "cache_event_time": 1750251842521445359,
+  "cache_state": "hit",
+  "compile_id": "0/0",
+  "components": [
+    "[b6z7ijvsnbtjurvqwdotlb6qtuyjsz4krtpkqj5thbq3fhqsbhp] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1, arg2_1):\n    scaled_fake_quantize = torch.ops.quark.scaled_fake_quantize.default('int4', arg0_1, arg1_1, arg2_1, 0, 0, 0.0, 15.0, 0, 'per_channel', 'haha');  arg0_1 = arg1_1 = arg2_1 = None\n    return (scaled_fake_quantize,)\n    \n# To see more debug info, please use `graph_module.print_readable()`",
+    "[nef3d47wx7hdxcyvwgucihijvag3qlbodqmms7dktjkxqgk4gux] example_inputs[0]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096, 8192]), stride=(8192, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+    "[7eps4cy57l62xkqx7rnf6ttfsuweegaqsqravzj2x5h4v6fbtod] example_inputs[1]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+    "[dpovg5nw4n7fkaetfrkwcqepypqcssr7fzpd75svn53n4xjxfow] example_inputs[2]: TensorMetadata(dtype=torch.int32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
+    "[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
+    "[xq2hdkbfkbcuye6rgtypayrkhqf4cntij2dsd24rei3lsknakkf] fx_kwargs[cudagraphs]: BoxedBool(value=False)",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
+    "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []",
+    "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inputs_to_check[1]: 1",
+    "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inputs_to_check[2]: 2",
+    "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
+    "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
+    "[ol6sj5vrtmgqd5cnfjbun7wbukfn7jyrtpkapnhmsysjdypit4i] torch_version: <bytes>",
+    "[rlwghyno43ot2zz35l3xlcrrh3arqenri6t5nsopt3t3q333zgy] system_info[device]: {'name': 'gfx942:sramecc+:xnack-'}",
+    "[njkbnmcun3rnt2yz2lgf57oprgdbjzlkadbcfoy57sir5fnskzg] system_info[version]: {'triton': '3.3.02e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-364a7d1dd5f29867a741138f83c5b453259e240a5d51f364f24f7196cedbf442-2e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ca6686d24a6f780b8449b43d419d11c978ebd00ab87a5fc6e57198a2027680d0-00deb4ba92653e089ad09c2f67cbc85602c84cd3ee2347ddcfcccc2081cfa45e-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-78995dd40c4e54964f62cdc6d47e66f6d9df8b55b172ad7eb99235f27999c840-f7c26e8ffdaf8cd2491de04217d32827b1a4f859f7f93ea56e26590c9f5e071a-a5b982237b29a230519ad566660a5912ae9d31da0dd6388465f3a1043b12ece7-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-f983f9d6d6f987af520297c8fe3185601ae8f7d60bacab880ac9326bdfee1f67-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-bd364752852b76a0b75a9d93ecbc239781b730aa75021445a02d795aa8d38f6a-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-b616015f724e553348f5b019f7324dec130f7bbaf984d43300fa69c7c2fdda2f-54fe722cbe379a55695ab9478e73d344377cf5e9d6e055aff7cd03bf6fff1b2a-10285555cd515e21ca54714fc8eb9c173cca6b092b5e951a17ae5eee28ed2707-f2d4e73182e68eddc6237577b2158b7d8498ccb3c50642b9c74c55d3f4be3943', 'hip': '6.3.42131-fa1d09cbd'}",
+    "[3qefpuc3yyxaxf25vrqt6skezzayzydwq2o6i2xf2qdyioam5zw] system_info[hash]: 6e7af1eb52fca9b11dcfccdeb7b2a74263a0c6d2c07fe2b7f2c2003a4663609b",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
+    "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
+    "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[online_softmax]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
+    "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
+    "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
+    "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
+    "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
+    "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
+    "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
+    "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
+    "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
+    "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[max_autotune]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
+    "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
+    "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
+    "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
+    "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
+    "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
+    "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
+    "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
+    "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[coordinate_descent_tuning]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
+    "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
+    "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
+    "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[layout_opt_default]: 0",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[layout_optimization]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
+    "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
+    "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
+    "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
+    "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
+    "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
+    "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
+    "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_nightly_or_source]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[developer_warnings]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
+    "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
+    "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
+    "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
+    "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
+    "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
+    "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
+    "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
+    "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
+    "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraphs]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
+    "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
+    "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
+    "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
+    "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
+    "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
+    "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
+    "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.disallow_failing_autotune_kernels_TESTING_ONLY]: False",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
+    "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
+    "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
+    "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
+    "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
+    "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
+    "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
+    "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
+    "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
+    "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
+    "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
+    "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
+    "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
+    "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.kBatch_sweep]: None",
+    "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[rocm.split_k_threshold]: 16",
+    "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
+    "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
+    "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
+    "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
+    "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
+    "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
+    "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
+    "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
+    "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
+  ],
+  "key": "fr57fwh6rfk65wtgsmaeq6p7domzwaxezpig6cnku7tl6zrhoujz",
+  "time_saved_ns": 427407491
+}

-_0_0_0/inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html

@@ -0,0 +1,67 @@
+<pre style="background-color:#ffffff;">
+<span style="font-style:italic;color:#969896;"># AOT ID: [&#39;0_inference&#39;]
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">ctypes </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">c_void_p, c_long, c_int
+</span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">torch
+</span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">math
+</span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">random
+</span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">os
+</span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">tempfile
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">math </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">inf, nan
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">cmath </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">nanj
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.hooks </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">run_intermediate_hooks
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.utils </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">maybe_profile
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.codegen.memory_planning </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">_align </span><span style="font-weight:bold;color:#a71d5d;">as </span><span style="color:#323232;">align
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">device, empty_strided
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.async_compile </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">AsyncCompile
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.select_algorithm </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">extern_kernels
+</span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.codegen.multi_kernel </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">MultiKernelCall
+</span><span style="color:#323232;">
+</span><span style="color:#323232;">aten </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.ops.aten
+</span><span style="color:#323232;">inductor_ops </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.ops.inductor
+</span><span style="color:#323232;">_quantized </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.ops._quantized
+</span><span style="color:#323232;">assert_size_stride </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._dynamo.guards.assert_size_stride
+</span><span style="color:#323232;">empty_strided_cpu </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._dynamo.guards._empty_strided_cpu
+</span><span style="color:#323232;">empty_strided_cuda </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._dynamo.guards._empty_strided_cuda
+</span><span style="color:#323232;">empty_strided_xpu </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._dynamo.guards._empty_strided_xpu
+</span><span style="color:#323232;">reinterpret_tensor </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._dynamo.guards._reinterpret_tensor
+</span><span style="color:#323232;">alloc_from_pool </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.ops.inductor._alloc_from_pool
+</span><span style="color:#323232;">async_compile </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">AsyncCompile()
+</span><span style="color:#323232;">empty_strided_p2p </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.</span><span style="color:#0086b3;">_C</span><span style="color:#323232;">._distributed_c10d._SymmetricMemory.empty_strided_p2p
+</span><span style="color:#323232;">
+</span><span style="color:#323232;">
+</span><span style="color:#323232;">async_compile.wait(</span><span style="color:#62a35c;">globals</span><span style="color:#323232;">())
+</span><span style="font-weight:bold;color:#a71d5d;">del </span><span style="color:#323232;">async_compile
+</span><span style="color:#323232;">
+</span><span style="font-weight:bold;color:#a71d5d;">def </span><span style="font-weight:bold;color:#323232;">call</span><span style="color:#323232;">(args):
+</span><span style="color:#323232;">    arg0_1, arg1_1, arg2_1 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">args
+</span><span style="color:#323232;">    args.clear()
+</span><span style="color:#323232;">    assert_size_stride(arg0_1, (</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, </span><span style="color:#0086b3;">8192</span><span style="color:#323232;">), (</span><span style="color:#0086b3;">8192</span><span style="color:#323232;">, </span><span style="color:#0086b3;">1</span><span style="color:#323232;">))
+</span><span style="color:#323232;">    assert_size_stride(arg1_1, (</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, ), (</span><span style="color:#0086b3;">1</span><span style="color:#323232;">, ))
+</span><span style="color:#323232;">    assert_size_stride(arg2_1, (</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, ), (</span><span style="color:#0086b3;">1</span><span style="color:#323232;">, ))
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">with </span><span style="color:#323232;">torch.cuda._DeviceGuard(</span><span style="color:#0086b3;">0</span><span style="color:#323232;">):
+</span><span style="color:#323232;">        torch.cuda.set_device(</span><span style="color:#0086b3;">0</span><span style="color:#323232;">)
+</span><span style="color:#323232;">        </span><span style="font-style:italic;color:#969896;"># Topologically Sorted Source Nodes: [scaled_fake_quantize], Original ATen: [quark.scaled_fake_quantize]
+</span><span style="color:#323232;">        buf0 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">torch.ops.quark.scaled_fake_quantize.default(</span><span style="color:#183691;">&#39;int4&#39;</span><span style="color:#323232;">, arg0_1, arg1_1, arg2_1, </span><span style="color:#0086b3;">0</span><span style="color:#323232;">, </span><span style="color:#0086b3;">0</span><span style="color:#323232;">, </span><span style="color:#0086b3;">0.0</span><span style="color:#323232;">, </span><span style="color:#0086b3;">15.0</span><span style="color:#323232;">, </span><span style="color:#0086b3;">0</span><span style="color:#323232;">, </span><span style="color:#183691;">&#39;per_channel&#39;</span><span style="color:#323232;">, </span><span style="color:#183691;">&#39;haha&#39;</span><span style="color:#323232;">)
+</span><span style="color:#323232;">        </span><span style="font-weight:bold;color:#a71d5d;">del </span><span style="color:#323232;">arg0_1
+</span><span style="color:#323232;">        </span><span style="font-weight:bold;color:#a71d5d;">del </span><span style="color:#323232;">arg1_1
+</span><span style="color:#323232;">        </span><span style="font-weight:bold;color:#a71d5d;">del </span><span style="color:#323232;">arg2_1
+</span><span style="color:#323232;">        buf1 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">buf0
+</span><span style="color:#323232;">        assert_size_stride(buf1, (</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, </span><span style="color:#0086b3;">8192</span><span style="color:#323232;">), (</span><span style="color:#0086b3;">8192</span><span style="color:#323232;">, </span><span style="color:#0086b3;">1</span><span style="color:#323232;">))
+</span><span style="color:#323232;">        </span><span style="font-weight:bold;color:#a71d5d;">del </span><span style="color:#323232;">buf0
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">return </span><span style="color:#323232;">(buf1, )
+</span><span style="color:#323232;">
+</span><span style="color:#323232;">
+</span><span style="font-weight:bold;color:#a71d5d;">def </span><span style="font-weight:bold;color:#323232;">benchmark_compiled_module</span><span style="color:#323232;">(times</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#0086b3;">10</span><span style="color:#323232;">, repeat</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#0086b3;">10</span><span style="color:#323232;">):
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._dynamo.testing </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">rand_strided
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.utils </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">print_performance
+</span><span style="color:#323232;">    arg0_1 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">rand_strided((</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, </span><span style="color:#0086b3;">8192</span><span style="color:#323232;">), (</span><span style="color:#0086b3;">8192</span><span style="color:#323232;">, </span><span style="color:#0086b3;">1</span><span style="color:#323232;">), device</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#183691;">&#39;cuda:0&#39;</span><span style="color:#323232;">, dtype</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#323232;">torch.float32)
+</span><span style="color:#323232;">    arg1_1 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">rand_strided((</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, ), (</span><span style="color:#0086b3;">1</span><span style="color:#323232;">, ), device</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#183691;">&#39;cuda:0&#39;</span><span style="color:#323232;">, dtype</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#323232;">torch.float32)
+</span><span style="color:#323232;">    arg2_1 </span><span style="font-weight:bold;color:#a71d5d;">= </span><span style="color:#323232;">rand_strided((</span><span style="color:#0086b3;">4096</span><span style="color:#323232;">, ), (</span><span style="color:#0086b3;">1</span><span style="color:#323232;">, ), device</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#183691;">&#39;cuda:0&#39;</span><span style="color:#323232;">, dtype</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#323232;">torch.int32)
+</span><span style="color:#323232;">    fn </span><span style="font-weight:bold;color:#a71d5d;">= lambda</span><span style="color:#323232;">: call([arg0_1, arg1_1, arg2_1])
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">return </span><span style="color:#323232;">print_performance(fn, times</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#323232;">times, repeat</span><span style="font-weight:bold;color:#a71d5d;">=</span><span style="color:#323232;">repeat)
+</span><span style="color:#323232;">
+</span><span style="color:#323232;">
+</span><span style="font-weight:bold;color:#a71d5d;">if </span><span style="color:#323232;">__name__ </span><span style="font-weight:bold;color:#a71d5d;">== </span><span style="color:#183691;">&quot;__main__&quot;</span><span style="color:#323232;">:
+</span><span style="color:#323232;">    </span><span style="font-weight:bold;color:#a71d5d;">from </span><span style="color:#323232;">torch._inductor.wrapper_benchmark </span><span style="font-weight:bold;color:#a71d5d;">import </span><span style="color:#323232;">compiled_module_main
+</span><span style="color:#323232;">    compiled_module_main(</span><span style="color:#183691;">&#39;None&#39;</span><span style="color:#323232;">, benchmark_compiled_module)
+</span></pre>

-_0_0_0/inductor_pre_grad_graph_1.txt

@@ -0,0 +1,15 @@
+class GraphModule(torch.nn.Module):
+    def forward(self, L_w_: "f32[4096, 8192][8192, 1]cuda:0", L_scale_: "f32[4096][1]cuda:0", L_zero_point_: "i32[4096][1]cuda:0"):
+        l_w_ = L_w_
+        l_scale_ = L_scale_
+        l_zero_point_ = L_zero_point_
+        
+         # File: /shared_volume/repos/quark/bench_qdq.py:8 in run_scaled_fake_quantize, code: return scaled_fake_quantize(quant_dtype, w, scale, zero_point, axis, group_size, quant_min, quant_max, round_mode, qscheme, mx_element_dtype)
+        function_ctx = torch.autograd.function.FunctionCtx();  function_ctx = None
+        
+         # File: /shared_volume/repos/quark/quark/torch/kernel/__init__.py:168 in forward, code: return ops.quark.scaled_fake_quantize(quant_dtype, inputs, scale, zero_point, axis, group_size, quant_min,
+        scaled_fake_quantize: "f32[4096, 8192][8192, 1]cuda:0" = torch.ops.quark.scaled_fake_quantize('int4', l_w_, l_scale_, l_zero_point_, 0, 0, 0, 15, 0, 'per_channel', 'haha');  l_w_ = l_scale_ = l_zero_point_ = None
+        return (scaled_fake_quantize,)
+        
+
+ # graph id: 140371115438448

-_0_0_0/torch._functorch.config_6.txt

@@ -0,0 +1,36 @@
+{
+"TYPE_CHECKING": false,
+"functionalize_rng_ops": false,
+"fake_tensor_allow_meta": true,
+"debug_assert": false,
+"debug_partitioner": false,
+"decompose_custom_triton_ops": true,
+"static_weight_shapes": true,
+"cse": true,
+"enable_autograd_cache": true,
+"enable_remote_autograd_cache": null,
+"view_replay_for_aliased_outputs": true,
+"max_dist_from_bw": 1000,
+"ban_recompute_used_far_apart": true,
+"ban_recompute_long_fusible_chains": true,
+"ban_recompute_materialized_backward": true,
+"ban_recompute_not_in_allowlist": true,
+"ban_recompute_reductions": true,
+"recompute_views": false,
+"activation_memory_budget": 1.0,
+"activation_memory_budget_runtime_estimator": "flops",
+"activation_memory_budget_solver": "dp",
+"visualize_memory_budget_pareto": false,
+"memory_budget_pareto_dir": null,
+"aggressive_recomputation": false,
+"fake_tensor_allow_unsafe_data_ptr_access": true,
+"unlift_effect_tokens": true,
+"fake_tensor_crossref": false,
+"fake_tensor_propagate_real_tensors": false,
+"donated_buffer": true,
+"torch_compile_graph_format": "svg",
+"generate_fake_kernels_from_real_mismatches": false,
+"graphsafe_rng_functionalization": true,
+"strict_autograd_cache": false,
+"disable_guess_zero_tangent_for_mutated_input_subclass": false
+}

chromium_events.json

@@ -0,0 +1,962 @@
+[
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "dynamo",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841454820.0
+  },
+  {
+    "args": {
+      "compile_id": "0/0",
+      "fn_name": "_compile.compile_inner"
+    },
+    "cat": "dynamo_timed",
+    "name": "entire_frame_compile",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841482125.8
+  },
+  {
+    "args": {
+      "compile_id": "0/0",
+      "fn_name": "OutputGraph.call_user_compiler"
+    },
+    "cat": "dynamo_timed",
+    "name": "backend_compile",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841551173.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "_recursive_pre_grad_passes",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841778018.5
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "_recursive_pre_grad_passes",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841791426.5
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "create_aot_dispatcher_function",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841793402.5
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "compile_fx.<locals>.fw_compiler_base",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841814442.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "_recursive_joint_graph_passes",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251841815353.0
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "_recursive_joint_graph_passes",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842111758.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0",
+      "fn_name": "compile_fx_inner"
+    },
+    "cat": "dynamo_timed",
+    "name": "inductor_compile",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842112457.0
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "inductor_codecache_torch_key",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842118406.0
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "inductor_codecache_torch_key",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842234516.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "PyCodeCache.load_by_key_path",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842504139.8
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "async_compile.wait",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842508906.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "async_compile.wait",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842509594.5
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "PyCodeCache.load_by_key_path",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842509931.5
+  },
+  {
+    "args": {
+      "cache_event_time": 1750251842521445359,
+      "cache_state": "hit",
+      "compile_id": "0/0",
+      "components": [
+        "[b6z7ijvsnbtjurvqwdotlb6qtuyjsz4krtpkqj5thbq3fhqsbhp] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1, arg2_1):\n    scaled_fake_quantize = torch.ops.quark.scaled_fake_quantize.default('int4', arg0_1, arg1_1, arg2_1, 0, 0, 0.0, 15.0, 0, 'per_channel', 'haha');  arg0_1 = arg1_1 = arg2_1 = None\n    return (scaled_fake_quantize,)\n    \n# To see more debug info, please use `graph_module.print_readable()`",
+        "[nef3d47wx7hdxcyvwgucihijvag3qlbodqmms7dktjkxqgk4gux] example_inputs[0]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096, 8192]), stride=(8192, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[7eps4cy57l62xkqx7rnf6ttfsuweegaqsqravzj2x5h4v6fbtod] example_inputs[1]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[dpovg5nw4n7fkaetfrkwcqepypqcssr7fzpd75svn53n4xjxfow] example_inputs[2]: TensorMetadata(dtype=torch.int32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
+        "[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
+        "[xq2hdkbfkbcuye6rgtypayrkhqf4cntij2dsd24rei3lsknakkf] fx_kwargs[cudagraphs]: BoxedBool(value=False)",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []",
+        "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inputs_to_check[1]: 1",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inputs_to_check[2]: 2",
+        "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
+        "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
+        "[ol6sj5vrtmgqd5cnfjbun7wbukfn7jyrtpkapnhmsysjdypit4i] torch_version: <bytes>",
+        "[rlwghyno43ot2zz35l3xlcrrh3arqenri6t5nsopt3t3q333zgy] system_info[device]: {'name': 'gfx942:sramecc+:xnack-'}",
+        "[njkbnmcun3rnt2yz2lgf57oprgdbjzlkadbcfoy57sir5fnskzg] system_info[version]: {'triton': '3.3.02e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-364a7d1dd5f29867a741138f83c5b453259e240a5d51f364f24f7196cedbf442-2e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ca6686d24a6f780b8449b43d419d11c978ebd00ab87a5fc6e57198a2027680d0-00deb4ba92653e089ad09c2f67cbc85602c84cd3ee2347ddcfcccc2081cfa45e-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-78995dd40c4e54964f62cdc6d47e66f6d9df8b55b172ad7eb99235f27999c840-f7c26e8ffdaf8cd2491de04217d32827b1a4f859f7f93ea56e26590c9f5e071a-a5b982237b29a230519ad566660a5912ae9d31da0dd6388465f3a1043b12ece7-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-f983f9d6d6f987af520297c8fe3185601ae8f7d60bacab880ac9326bdfee1f67-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-bd364752852b76a0b75a9d93ecbc239781b730aa75021445a02d795aa8d38f6a-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-b616015f724e553348f5b019f7324dec130f7bbaf984d43300fa69c7c2fdda2f-54fe722cbe379a55695ab9478e73d344377cf5e9d6e055aff7cd03bf6fff1b2a-10285555cd515e21ca54714fc8eb9c173cca6b092b5e951a17ae5eee28ed2707-f2d4e73182e68eddc6237577b2158b7d8498ccb3c50642b9c74c55d3f4be3943', 'hip': '6.3.42131-fa1d09cbd'}",
+        "[3qefpuc3yyxaxf25vrqt6skezzayzydwq2o6i2xf2qdyioam5zw] system_info[hash]: 6e7af1eb52fca9b11dcfccdeb7b2a74263a0c6d2c07fe2b7f2c2003a4663609b",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
+        "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
+        "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[online_softmax]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
+        "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
+        "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
+        "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
+        "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
+        "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
+        "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
+        "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[max_autotune]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
+        "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
+        "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
+        "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
+        "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
+        "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
+        "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
+        "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[coordinate_descent_tuning]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
+        "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
+        "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[layout_opt_default]: 0",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[layout_optimization]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
+        "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
+        "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
+        "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_nightly_or_source]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[developer_warnings]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
+        "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
+        "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
+        "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
+        "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
+        "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
+        "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
+        "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraphs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
+        "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
+        "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
+        "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
+        "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
+        "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.disallow_failing_autotune_kernels_TESTING_ONLY]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
+        "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
+        "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
+        "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
+        "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.kBatch_sweep]: None",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[rocm.split_k_threshold]: 16",
+        "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
+        "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
+        "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
+        "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
+        "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
+        "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
+      ],
+      "key": "fr57fwh6rfk65wtgsmaeq6p7domzwaxezpig6cnku7tl6zrhoujz",
+      "time_saved_ns": 427407491
+    },
+    "cat": "dynamo_timed",
+    "name": "fx_graph_cache_hit",
+    "ph": "i",
+    "pid": 0,
+    "s": "p",
+    "tid": 0,
+    "ts": 1750251842118276.5
+  },
+  {
+    "args": {
+      "cache_bypass_reason": null,
+      "cache_event_time": 1750251842118276381,
+      "cache_state": "hit",
+      "compile_id": "0/0",
+      "components": [
+        "[b6z7ijvsnbtjurvqwdotlb6qtuyjsz4krtpkqj5thbq3fhqsbhp] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1, arg2_1):\n    scaled_fake_quantize = torch.ops.quark.scaled_fake_quantize.default('int4', arg0_1, arg1_1, arg2_1, 0, 0, 0.0, 15.0, 0, 'per_channel', 'haha');  arg0_1 = arg1_1 = arg2_1 = None\n    return (scaled_fake_quantize,)\n    \n# To see more debug info, please use `graph_module.print_readable()`",
+        "[nef3d47wx7hdxcyvwgucihijvag3qlbodqmms7dktjkxqgk4gux] example_inputs[0]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096, 8192]), stride=(8192, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[7eps4cy57l62xkqx7rnf6ttfsuweegaqsqravzj2x5h4v6fbtod] example_inputs[1]: TensorMetadata(dtype=torch.float32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[dpovg5nw4n7fkaetfrkwcqepypqcssr7fzpd75svn53n4xjxfow] example_inputs[2]: TensorMetadata(dtype=torch.int32, shape=torch.Size([4096]), stride=(1,), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
+        "[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
+        "[xq2hdkbfkbcuye6rgtypayrkhqf4cntij2dsd24rei3lsknakkf] fx_kwargs[cudagraphs]: BoxedBool(value=False)",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []",
+        "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inputs_to_check[1]: 1",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inputs_to_check[2]: 2",
+        "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
+        "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
+        "[ol6sj5vrtmgqd5cnfjbun7wbukfn7jyrtpkapnhmsysjdypit4i] torch_version: <bytes>",
+        "[rlwghyno43ot2zz35l3xlcrrh3arqenri6t5nsopt3t3q333zgy] system_info[device]: {'name': 'gfx942:sramecc+:xnack-'}",
+        "[njkbnmcun3rnt2yz2lgf57oprgdbjzlkadbcfoy57sir5fnskzg] system_info[version]: {'triton': '3.3.02e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-364a7d1dd5f29867a741138f83c5b453259e240a5d51f364f24f7196cedbf442-2e234c1d93a7cae949341b60b8567f825914128ed598e1d846c8bc19a9d65ad8-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ca6686d24a6f780b8449b43d419d11c978ebd00ab87a5fc6e57198a2027680d0-00deb4ba92653e089ad09c2f67cbc85602c84cd3ee2347ddcfcccc2081cfa45e-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-78995dd40c4e54964f62cdc6d47e66f6d9df8b55b172ad7eb99235f27999c840-f7c26e8ffdaf8cd2491de04217d32827b1a4f859f7f93ea56e26590c9f5e071a-a5b982237b29a230519ad566660a5912ae9d31da0dd6388465f3a1043b12ece7-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-f983f9d6d6f987af520297c8fe3185601ae8f7d60bacab880ac9326bdfee1f67-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-bd364752852b76a0b75a9d93ecbc239781b730aa75021445a02d795aa8d38f6a-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-b616015f724e553348f5b019f7324dec130f7bbaf984d43300fa69c7c2fdda2f-54fe722cbe379a55695ab9478e73d344377cf5e9d6e055aff7cd03bf6fff1b2a-10285555cd515e21ca54714fc8eb9c173cca6b092b5e951a17ae5eee28ed2707-f2d4e73182e68eddc6237577b2158b7d8498ccb3c50642b9c74c55d3f4be3943', 'hip': '6.3.42131-fa1d09cbd'}",
+        "[3qefpuc3yyxaxf25vrqt6skezzayzydwq2o6i2xf2qdyioam5zw] system_info[hash]: 6e7af1eb52fca9b11dcfccdeb7b2a74263a0c6d2c07fe2b7f2c2003a4663609b",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
+        "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
+        "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[online_softmax]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
+        "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
+        "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
+        "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
+        "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
+        "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
+        "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
+        "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[max_autotune]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
+        "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
+        "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
+        "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
+        "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
+        "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
+        "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
+        "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[coordinate_descent_tuning]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
+        "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
+        "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[layout_opt_default]: 0",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[layout_optimization]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
+        "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
+        "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
+        "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
+        "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_nightly_or_source]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[developer_warnings]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
+        "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
+        "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
+        "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
+        "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
+        "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
+        "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
+        "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
+        "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraphs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
+        "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
+        "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
+        "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
+        "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
+        "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.disallow_failing_autotune_kernels_TESTING_ONLY]: False",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
+        "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
+        "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
+        "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
+        "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
+        "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
+        "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
+        "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
+        "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
+        "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
+        "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.kBatch_sweep]: None",
+        "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[rocm.split_k_threshold]: 16",
+        "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
+        "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
+        "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
+        "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
+        "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
+        "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
+        "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
+        "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
+        "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
+      ],
+      "fn_name": "compile_fx_inner",
+      "is_backward": false,
+      "key": "fr57fwh6rfk65wtgsmaeq6p7domzwaxezpig6cnku7tl6zrhoujz",
+      "local_cache_enabled": true,
+      "remote_cache_enabled": false
+    },
+    "cat": "dynamo_timed",
+    "name": "inductor_compile",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842527055.8
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "compile_fx.<locals>.fw_compiler_base",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842528002.0
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "create_aot_dispatcher_function",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842530288.2
+  },
+  {
+    "args": {
+      "cache_bypass_hard_exception": false,
+      "cache_bypass_reason": "Unsupported call_function target <class 'torch.autograd.function.FunctionCtx'>. \n Function module: torch.autograd.function, \nFunction name: FunctionCtx",
+      "cache_state": "bypass",
+      "compile_id": "0/0",
+      "components": [],
+      "key": null
+    },
+    "cat": "dynamo_timed",
+    "name": "autograd_cache_bypass",
+    "ph": "i",
+    "pid": 0,
+    "s": "p",
+    "tid": 0,
+    "ts": 1750251841793382.2
+  },
+  {
+    "args": {
+      "cache_bypass_reason": "Unsupported call_function target <class 'torch.autograd.function.FunctionCtx'>. \n Function module: torch.autograd.function, \nFunction name: FunctionCtx",
+      "cache_event_time": 1750251841793382207,
+      "cache_state": "bypass",
+      "compile_id": "0/0",
+      "components": [],
+      "dispatch_mode": "inference",
+      "fn_name": "OutputGraph.call_user_compiler",
+      "key": null,
+      "local_cache_enabled": true,
+      "remote_cache_enabled": false,
+      "requires_subclass_dispatch": false
+    },
+    "cat": "dynamo_timed",
+    "name": "backend_compile",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842544082.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0",
+      "fn_name": "_compile.compile_inner"
+    },
+    "cat": "dynamo_timed",
+    "name": "entire_frame_compile",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842551017.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "gc",
+    "ph": "B",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842551354.2
+  },
+  {
+    "args": {
+      "compile_id": "0/0"
+    },
+    "cat": "dynamo_timed",
+    "name": "gc",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842552469.2
+  },
+  {
+    "args": {
+      "accumulated_cache_size": 0,
+      "cache_size": 0,
+      "co_filename": "/shared_volume/repos/quark/bench_qdq.py",
+      "co_firstlineno": 7,
+      "co_name": "run_scaled_fake_quantize",
+      "compile_id": "0/0",
+      "compliant_custom_ops": [],
+      "distributed_ephemeral_timeout_us": 427407,
+      "dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": false, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}",
+      "dynamo_time_before_restart_s": 0.0,
+      "fail_reason": null,
+      "fail_type": null,
+      "fail_user_frame_filename": null,
+      "fail_user_frame_lineno": null,
+      "frame_key": "1",
+      "graph_input_count": 3,
+      "graph_node_count": 6,
+      "graph_op_count": 2,
+      "guard_count": 21,
+      "guard_latency_us": 0,
+      "has_guarded_code": true,
+      "non_compliant_ops": [
+        "quark::scaled_fake_quantize"
+      ],
+      "restart_reasons": [],
+      "shape_env_guard_count": 0
+    },
+    "cat": "dynamo_timed",
+    "name": "dynamo",
+    "ph": "E",
+    "pid": 0,
+    "tid": 0,
+    "ts": 1750251842558796.5
+  }
+]

compile_directory.json

@@ -0,0 +1,78 @@
+{
+  "[0/0]": {
+    "artifacts": [
+      {
+        "name": "dynamo_output_graph_0.txt",
+        "number": 0,
+        "suffix": "",
+        "url": "-_0_0_0/dynamo_output_graph_0.txt"
+      },
+      {
+        "name": "inductor_pre_grad_graph_1.txt",
+        "number": 1,
+        "suffix": "",
+        "url": "-_0_0_0/inductor_pre_grad_graph_1.txt"
+      },
+      {
+        "name": "before_recompile_pre_grad_2.txt",
+        "number": 2,
+        "suffix": "",
+        "url": "-_0_0_0/before_recompile_pre_grad_2.txt"
+      },
+      {
+        "name": "after_recompile_pre_grad_3.txt",
+        "number": 3,
+        "suffix": "",
+        "url": "-_0_0_0/after_recompile_pre_grad_3.txt"
+      },
+      {
+        "name": "aot_forward_graph_fw_metadata_4.txt",
+        "number": 4,
+        "suffix": "",
+        "url": "-_0_0_0/aot_forward_graph_fw_metadata_4.txt"
+      },
+      {
+        "name": "aot_inference_graph_5.txt",
+        "number": 5,
+        "suffix": "",
+        "url": "-_0_0_0/aot_inference_graph_5.txt"
+      },
+      {
+        "name": "torch._functorch.config_6.txt",
+        "number": 6,
+        "suffix": "",
+        "url": "-_0_0_0/torch._functorch.config_6.txt"
+      },
+      {
+        "name": "inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html",
+        "number": 7,
+        "suffix": "",
+        "url": "-_0_0_0/inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html"
+      },
+      {
+        "name": "fx_graph_cache_hit_8.json",
+        "number": 8,
+        "suffix": "✅",
+        "url": "-_0_0_0/fx_graph_cache_hit_8.json"
+      },
+      {
+        "name": "aotautograd_cache_bypass_9.json",
+        "number": 9,
+        "suffix": "❓",
+        "url": "-_0_0_0/aotautograd_cache_bypass_9.json"
+      },
+      {
+        "name": "dynamo_cpp_guards_str_10.txt",
+        "number": 10,
+        "suffix": "",
+        "url": "-_0_0_0/dynamo_cpp_guards_str_10.txt"
+      },
+      {
+        "name": "compilation_metrics_11.html",
+        "number": 11,
+        "suffix": "",
+        "url": "-_0_0_0/compilation_metrics_11.html"
+      }
+    ]
+  }
+}

failures_and_restarts.html

@@ -0,0 +1,72 @@
+
+<html>
+<head>
+    <style>
+    
+table {
+    width: 90%;
+    border-collapse: collapse;
+    margin: 20px 0;
+}
+table, th, td {
+    border: 1px solid #999;
+    padding: 10px;
+    text-align: left;
+}
+th {
+    background-color: #d3d3d3;
+    font-weight: bold;
+}
+tr:nth-child(odd) {
+    background-color: #f2f2f2;
+}
+a {
+    color: #0066cc;
+    text-decoration: none;
+}
+a:hover {
+    text-decoration: underline;
+}
+
+    </style>
+</head>
+<body>
+    <h1>Failures and Restarts</h1>
+    <table>
+    <tr> <th> Compile Id </th> <th> Failure Type </th> <th> Failure Description </th> <th> Failure Source (compilation failures only) </th> </tr>
+    
+    
+    <script>
+    document.addEventListener('DOMContentLoaded', function() {
+
+        // Append the current URL's query parameters to all relative links on the page
+        const queryParams = new URLSearchParams(window.location.search);
+        if (queryParams.size === 0) return url; // No query params, return original URL
+
+        function appendQueryParams(url) {
+            const newURL = new URL((new Request(url)).url);  // new URL(<relative URL>) but it actually works
+            const newSearchParams = new URLSearchParams(newURL.searchParams);
+            console.log(newURL.searchParams);
+            console.log(newSearchParams);
+
+            // Append query parameters
+            for (const [key, value] of queryParams) {
+                newSearchParams.set(key, value);
+            }
+
+            newURL.search = newSearchParams;
+            return newURL;
+        }
+
+        // Select all relative links on the page
+        const relativeLinks = document.querySelectorAll('a[href]:not([href^="http://"]):not([href^="https://"]):not([href^="\#"])');
+
+        // Append query parameters to each relative link
+        relativeLinks.forEach((link) => {
+            link.setAttribute("href", appendQueryParams(link.getAttribute("href")))
+        });
+    });
+    </script>
+
+</body>
+</html>

index.html

@@ -0,0 +1,202 @@
+
+<html>
+<head>
+  <meta charset="UTF-8">
+</head>
+<style>
+
+table td { vertical-align: top; }
+
+.stack-trie { white-space: nowrap; font-family: monospace; }
+.stack-trie ul { padding-left: 1ch;  }
+.stack-trie li { margin-left: 1ch; list-style-type: none;  }
+.stack-trie .marker {
+  cursor: pointer;
+}
+.stack-trie .marker.collapsed::before {
+  content: "+ ";
+}
+.stack-trie .marker:not(.collapsed)::before {
+  content: "- ";
+}
+.stack-trie a { text-decoration: none; }
+.stack-trie a:hover { text-decoration: underline; }
+.status-missing { background-color: purple; color: white; }
+.status-error { background-color: red; color: white; }
+.status-empty { background-color: white; color: black; }
+.status-ok { background-color: green; color: white; }
+.status-break { background-color: lime; color: black; }
+summary::-webkit-details-marker { color: #00ACF3; font-size: 125%; margin-right: 2px; }
+summary:focus { outline-style: none; }
+article > details > summary { font-size: 28px; margin-top: 16px; }
+details > p { margin-left: 24px; }
+details details summary { font-size: 16px; }
+
+</style>
+<script>
+
+  function toggleList(toggleItem) {
+    const listItem = toggleItem.parentNode;
+    const nestedList = listItem.querySelector('ul');
+    if (nestedList) {
+      nestedList.style.display = nestedList.style.display === 'none' ? 'block' : 'none';
+
+      // Toggle the collapse/expand indicator
+      toggleItem.classList.toggle('collapsed');
+    }
+  }
+
+</script>
+<body>
+<div>
+
+<h2>Stack trie</h2>
+<p>
+The <strong>stack trie</strong> is a way of getting a quick orientation on where all the
+compilations in a model take place, esp., if you are compiling a codebase you are unfamiliar with.
+It is a tree of stack frames, for all stacks that triggered PT2 compilation.  If only a single
+stack is in the tree, you will simply see a plain list of frames (most recent call last).  With
+multiple stacks, at every point where two stacks diverge from having a common prefix, we increase
+the indentation of the list and have a separate sub-list per sub-tree.
+</p>
+<p>
+Links to particular compilation are color coded by status:
+<span class="status-ok">[Success]</span>,
+<span class="status-break">[Success with restart (e.g., graph break)]</span>,
+<span class="status-empty">[Empty graph]</span>,
+<span class="status-error">[Error]</span>,
+<span class="status-missing">[Metrics were missing]</span>
+</p>
+<details><summary>Stack</summary><div class='stack-trie'><ul><li>/shared_volume/repos/quark/bench_qdq.py:161 in &lt;module&gt;<br>&nbsp;&nbsp;&nbsp;&nbsp;mean, median = do_bench(run_scaled_fake_quantize_comp, kwargs_scaled_fake_quantize, num_runs=num_runs, num_warmup=num_warmup, name="quark qdq")</li>
+<li>/shared_volume/repos/quark/bench_qdq.py:70 in do_bench<br>&nbsp;&nbsp;&nbsp;&nbsp;f(**kwargs)</li>
+<li><a href='#[0/0]' class='status-ok'>[0/0]</a> /shared_volume/repos/quark/bench_qdq.py:7 in run_scaled_fake_quantize<br>&nbsp;&nbsp;&nbsp;&nbsp;</li>
+</ul></div></details>
+</div>
+<div>
+
+<h2>IR dumps</h2>
+<p>
+The <strong>IR dumps</strong> collected dumped intermediate products from various points of the PT2
+compilation process.  The products are organized by compile id, and then sorted in chronological
+order.
+</p>
+<p>
+A <strong>compile id</strong> uniquely identifies are particular compilation inside a PT2
+program.  It is traditionally written as <code>[x/y]</code>, where the <strong>frame id</strong> x
+identifies the particular Python frame which we are compiling, and <strong>frame compile
+id</strong> y identifies how many times we've recompiled this same frame.  For example,
+<code>[0/0]</code> refers to the very first frame compiled by PT2; <code>[0/1]</code> refers to the
+first recompilation of this frame, while <code>[1/0]</code> refers to a different frame, within
+distinct code cache, which we are compiling next (perhaps because of a graph break).  Although
+Dynamo treats distinct frames as completely unrelated, a frame compilation could overlap with another
+frame; for example, if you graph break in an inlined function, Dynamo will typically try to compile
+the nested frame again on an inner frame.  You can identify the hierarchical relationship between
+frames by looking at the stack trie above.
+</p>
+<p>
+In some situations, the compile id will have an extra signifier <code>[x/y_z]</code>, where z is the
+<strong>attempt</strong> for this particular (re)compilation.  Certain conditions will cause Dynamo to
+restart analysis, when Dynamo discovers that it needs to undo a decision it previously made.  The most
+common cause of recompilation is a graph break in an inlined function call, which forces to restart
+and avoid inlining the function in the first place.
+</p>
+<p>
+When compiled autograd is enabled, the compile id will include a prefix signifier <code>[!a/x/y]</code>,
+where a is the <strong>compiled autograd id</strong>. For instance, <code>[!0/-/-]</code> refers 
+to the first graph captured by compiled autograd. It is then traced by torch.compile as <code>[!0/x/y_z]</code>.
+</p>
+<p>
+Here is a high level description of PT2's compilation phases, and the intermediate products each
+phase generates:
+</p>
+<ol>
+<li><em>Optional:</em> If compiled autograd is enabled, and we are processing a backward call, compiled autograd will trace the autograd graph from the autograd engine, and produce an FX graph <code>compiled_autograd_graph</code> that will be Dynamo traced.  Otherwise, Dynamo will directly trace user's bytecode.</li>
+<li>Dynamo symbolically evaluates the Python bytecode of a program, producing <code>dynamo_output_graph</code></li>
+<li><em>Optional:</em> If <code>optimize_ddp</code> is enabled, the DDPOptimizer will split the Dynamo output graph to improve pipelining communications.  Each split subgraph is <code>optimize_ddp_split_child_submod</code>, and the high level graph that plumbs the graphs together is <code>optimize_ddp_split_graph</code>.  If there are multiple splits, each subsequent build product will be produced multiple times, one for each split.</li>
+<li>AOTAutograd traces the (possibly split) Dynamo output graph, producing a <code>aot_joint_graph</code> if backwards is enabled.  It then partitions the graph into <code>aot_forward_graph</code> and <code>aot_backward_graph</code>.  If training is not needed, there may only be an <code>aot_inference_graph</code>.</li>
+<li>Inductor will apply some post grad FX passes, producing <code>inductor_post_grad_graph</code></li>
+<li>Inductor will perform code generation, producing the final <code>inductor_output_code</code> which will be executed at runtime.  This output is a valid Python program and can be directly run.</li>
+</ol>
+
+
+<h2> Chromium Events </h2>
+PT2 generates <a href='chromium_events.json'>Chromium Trace Events</a> in JSON on specific events during compilation.
+You can download and view them in a tool like <a href='https://ui.perfetto.dev/'>Perfetto</a>.
+
+<p>
+Build products below:
+</p>
+<ul>
+
+    <li><a id="[0/0]">[0/0]</a>
+    <ul>
+        
+            <li><a href="-_0_0_0/dynamo_output_graph_0.txt">-_0_0_0/dynamo_output_graph_0.txt</a>  (0)</li>
+        
+            <li><a href="-_0_0_0/inductor_pre_grad_graph_1.txt">-_0_0_0/inductor_pre_grad_graph_1.txt</a>  (1)</li>
+        
+            <li><a href="-_0_0_0/before_recompile_pre_grad_2.txt">-_0_0_0/before_recompile_pre_grad_2.txt</a>  (2)</li>
+        
+            <li><a href="-_0_0_0/after_recompile_pre_grad_3.txt">-_0_0_0/after_recompile_pre_grad_3.txt</a>  (3)</li>
+        
+            <li><a href="-_0_0_0/aot_forward_graph_fw_metadata_4.txt">-_0_0_0/aot_forward_graph_fw_metadata_4.txt</a>  (4)</li>
+        
+            <li><a href="-_0_0_0/aot_inference_graph_5.txt">-_0_0_0/aot_inference_graph_5.txt</a>  (5)</li>
+        
+            <li><a href="-_0_0_0/torch._functorch.config_6.txt">-_0_0_0/torch._functorch.config_6.txt</a>  (6)</li>
+        
+            <li><a href="-_0_0_0/inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html">-_0_0_0/inductor_output_code_ch44xxkifazlcpkp6mi44xhqeej2j5mbgwmesiwx6y3oajzmixxp_7.html</a>  (7)</li>
+        
+            <li><a href="-_0_0_0/fx_graph_cache_hit_8.json">-_0_0_0/fx_graph_cache_hit_8.json</a> ✅ (8)</li>
+        
+            <li><a href="-_0_0_0/aotautograd_cache_bypass_9.json">-_0_0_0/aotautograd_cache_bypass_9.json</a> ❓ (9)</li>
+        
+            <li><a href="-_0_0_0/dynamo_cpp_guards_str_10.txt">-_0_0_0/dynamo_cpp_guards_str_10.txt</a>  (10)</li>
+        
+            <li><a href="-_0_0_0/compilation_metrics_11.html">-_0_0_0/compilation_metrics_11.html</a>  (11)</li>
+        
+    </ul>
+    </li>
+
+</ul>
+</div>
+
+
+
+
+
+
+    <script>
+    document.addEventListener('DOMContentLoaded', function() {
+
+        // Append the current URL's query parameters to all relative links on the page
+        const queryParams = new URLSearchParams(window.location.search);
+        if (queryParams.size === 0) return url; // No query params, return original URL
+
+        function appendQueryParams(url) {
+            const newURL = new URL((new Request(url)).url);  // new URL(<relative URL>) but it actually works
+            const newSearchParams = new URLSearchParams(newURL.searchParams);
+            console.log(newURL.searchParams);
+            console.log(newSearchParams);
+
+            // Append query parameters
+            for (const [key, value] of queryParams) {
+                newSearchParams.set(key, value);
+            }
+
+            newURL.search = newSearchParams;
+            return newURL;
+        }
+
+        // Select all relative links on the page
+        const relativeLinks = document.querySelectorAll('a[href]:not([href^="http://"]):not([href^="https://"]):not([href^="\#"])');
+
+        // Append query parameters to each relative link
+        relativeLinks.forEach((link) => {
+            link.setAttribute("href", appendQueryParams(link.getAttribute("href")))
+        });
+    });
+    </script>
+
+</body>
+</html>