8bbf2c609c4768deaaa7e385ebcdbc327105a3aee3e91075f965c1e05d8c9244

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/tmp/tmpr864kgol",
+  "_name_or_path": "/tmp/tmpsnmqz3b1",
   "architectures": [
     "LlamaForCausalLM"
   ],

results.json

@@ -1,6 +1,26 @@
 {
     "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
     "base_current_gpu_total_memory": 40339.3125,
+    "base_token_generation_latency_sync": 83.62196044921875,
+    "base_token_generation_latency_async": 83.07278789579868,
+    "base_token_generation_throughput_sync": 0.011958581150549224,
+    "base_token_generation_throughput_async": 0.012037636214332154,
+    "base_token_generation_CO2_emissions": 2.0762580172851258e-05,
+    "base_token_generation_energy_consumption": 0.0068736865279905605,
+    "base_inference_latency_sync": 82.20262451171875,
+    "base_inference_latency_async": 79.88781929016113,
+    "base_inference_throughput_sync": 0.012165061711106812,
+    "base_inference_throughput_async": 0.012517552849551352,
     "smashed_current_gpu_type": "NVIDIA A100-PCIE-40GB",
-    "smashed_current_gpu_total_memory": 40339.3125
+    "smashed_current_gpu_total_memory": 40339.3125,
+    "smashed_token_generation_latency_sync": 177.8312194824219,
+    "smashed_token_generation_latency_async": 178.8266021758318,
+    "smashed_token_generation_throughput_sync": 0.0056233095792206905,
+    "smashed_token_generation_throughput_async": 0.005592009174433382,
+    "smashed_token_generation_CO2_emissions": 5.6607409622391035e-05,
+    "smashed_token_generation_energy_consumption": 0.019573642151998485,
+    "smashed_inference_latency_sync": 185.82732849121095,
+    "smashed_inference_latency_async": 144.77639198303223,
+    "smashed_inference_throughput_sync": 0.0053813398067943325,
+    "smashed_inference_throughput_async": 0.0069072034901740045
 }

smash_config.json

@@ -14,7 +14,7 @@
         "controlnet": "None",
         "unet_dim": 4,
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelst5pb_r7v",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelstj7ks5m6",
         "batch_size": 1,
         "model_name": "mattshumer/Llama-3-8B-16K",
         "task": "text_text_generation",