Upload cuda_training_transformers_text-generation_hf-internal-testing/tiny-random-LlamaForCausalLM/benchmark.json with huggingface_hub

cuda_training_transformers_text-generation_hf-internal-testing/tiny-random-LlamaForCausalLM/benchmark.json

@@ -3,7 +3,7 @@
         "name": "cuda_training_transformers_text-generation_hf-internal-testing/tiny-random-LlamaForCausalLM",
         "backend": {
             "name": "pytorch",
-            "version": "2.7.0+rocm6.3",
+            "version": "2.3.1+rocm5.7",
             "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
             "model": "hf-internal-testing/tiny-random-LlamaForCausalLM",
             "processor": "hf-internal-testing/tiny-random-LlamaForCausalLM",
@@ -11,7 +11,7 @@
             "library": "transformers",
             "model_type": "llama",
             "device": "cuda",
-            "device_ids": "5",
+            "device_ids": "6",
             "seed": 42,
             "inter_op_num_threads": null,
             "intra_op_num_threads": null,
@@ -109,7 +109,7 @@
             "optimum_commit": null,
             "timm_version": "1.0.15",
             "timm_commit": null,
-            "peft_version": null,
+            "peft_version": "0.15.2",
             "peft_commit": null
         },
         "print_report": true,
@@ -119,34 +119,34 @@
         "overall": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 2345.623552,
+                "max_ram": 1319.739392,
                 "max_global_vram": 68702.69952,
-                "max_process_vram": 0.0,
-                "max_reserved": 182.452224,
-                "max_allocated": 98.354688
+                "max_process_vram": 68802.019328,
+                "max_reserved": 44.040192,
+                "max_allocated": 25.311744
             },
             "latency": {
                 "unit": "s",
                 "values": [
-                    1.058366943359375,
-                    0.010661576271057129,
-                    0.009537578582763672,
-                    0.00980013942718506,
-                    0.009378539085388184
+                    0.7397705688476562,
+                    0.010124991416931153,
+                    0.00866274642944336,
+                    0.008307865142822266,
+                    0.008410265922546387
                 ],
                 "count": 5,
-                "total": 1.097744776725769,
-                "mean": 0.2195489553451538,
-                "p50": 0.00980013942718506,
-                "p90": 0.6392847965240479,
-                "p95": 0.8488258699417113,
-                "p99": 1.0164587286758422,
-                "stdev": 0.4194092278841988,
-                "stdev_": 191.03221294077392
+                "total": 0.7752764377593993,
+                "mean": 0.15505528755187986,
+                "p50": 0.00866274642944336,
+                "p90": 0.4479123378753663,
+                "p95": 0.5938414533615111,
+                "p99": 0.7105847457504272,
+                "stdev": 0.29235837438569023,
+                "stdev_": 188.55105104872365
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 45.547927952009424
+                "value": 64.49312472916543
             },
             "energy": null,
             "efficiency": null
@@ -154,31 +154,31 @@
         "warmup": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 2345.623552,
+                "max_ram": 1319.739392,
                 "max_global_vram": 68702.69952,
-                "max_process_vram": 0.0,
-                "max_reserved": 182.452224,
-                "max_allocated": 98.354688
+                "max_process_vram": 68802.019328,
+                "max_reserved": 44.040192,
+                "max_allocated": 25.311744
             },
             "latency": {
                 "unit": "s",
                 "values": [
-                    1.058366943359375,
-                    0.010661576271057129
+                    0.7397705688476562,
+                    0.010124991416931153
                 ],
                 "count": 2,
-                "total": 1.069028519630432,
-                "mean": 0.534514259815216,
-                "p50": 0.534514259815216,
-                "p90": 0.9535964066505432,
-                "p95": 1.005981675004959,
-                "p99": 1.0478898896884918,
-                "stdev": 0.5238526835441589,
-                "stdev_": 98.00537103074801
+                "total": 0.7498955602645874,
+                "mean": 0.3749477801322937,
+                "p50": 0.3749477801322937,
+                "p90": 0.6668060111045837,
+                "p95": 0.7032882899761199,
+                "p99": 0.732474113073349,
+                "stdev": 0.36482278871536256,
+                "stdev_": 97.2996262537256
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 7.48342991145422
+                "value": 10.668152238662863
             },
             "energy": null,
             "efficiency": null
@@ -186,32 +186,32 @@
         "train": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 2345.623552,
+                "max_ram": 1319.739392,
                 "max_global_vram": 68702.69952,
-                "max_process_vram": 0.0,
-                "max_reserved": 182.452224,
-                "max_allocated": 98.354688
+                "max_process_vram": 68802.019328,
+                "max_reserved": 44.040192,
+                "max_allocated": 25.311744
             },
             "latency": {
                 "unit": "s",
                 "values": [
-                    0.009537578582763672,
-                    0.00980013942718506,
-                    0.009378539085388184
+                    0.00866274642944336,
+                    0.008307865142822266,
+                    0.008410265922546387
                 ],
                 "count": 3,
-                "total": 0.028716257095336917,
-                "mean": 0.00957208569844564,
-                "p50": 0.009537578582763672,
-                "p90": 0.009747627258300782,
-                "p95": 0.00977388334274292,
-                "p99": 0.00979488821029663,
-                "stdev": 0.00017383856078811917,
-                "stdev_": 1.8160990850337657
+                "total": 0.025380877494812014,
+                "mean": 0.00846029249827067,
+                "p50": 0.008410265922546387,
+                "p90": 0.008612250328063965,
+                "p95": 0.008637498378753663,
+                "p99": 0.00865769681930542,
+                "stdev": 0.0001491356778406473,
+                "stdev_": 1.762772124854211
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 626.8226370951013
+                "value": 709.1953382494083
             },
             "energy": null,
             "efficiency": null