sharpenb commited on
Commit
9197534
1 Parent(s): ac222ed

8bbf2c609c4768deaaa7e385ebcdbc327105a3aee3e91075f965c1e05d8c9244

Browse files
Files changed (4) hide show
  1. config.json +1 -1
  2. plots.png +0 -0
  3. results.json +21 -1
  4. smash_config.json +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/tmp/tmpr864kgol",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
1
  {
2
+ "_name_or_path": "/tmp/tmpsnmqz3b1",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
plots.png ADDED
results.json CHANGED
@@ -1,6 +1,26 @@
1
  {
2
  "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
3
  "base_current_gpu_total_memory": 40339.3125,
 
 
 
 
 
 
 
 
 
 
4
  "smashed_current_gpu_type": "NVIDIA A100-PCIE-40GB",
5
- "smashed_current_gpu_total_memory": 40339.3125
 
 
 
 
 
 
 
 
 
 
6
  }
 
1
  {
2
  "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
3
  "base_current_gpu_total_memory": 40339.3125,
4
+ "base_token_generation_latency_sync": 83.62196044921875,
5
+ "base_token_generation_latency_async": 83.07278789579868,
6
+ "base_token_generation_throughput_sync": 0.011958581150549224,
7
+ "base_token_generation_throughput_async": 0.012037636214332154,
8
+ "base_token_generation_CO2_emissions": 2.0762580172851258e-05,
9
+ "base_token_generation_energy_consumption": 0.0068736865279905605,
10
+ "base_inference_latency_sync": 82.20262451171875,
11
+ "base_inference_latency_async": 79.88781929016113,
12
+ "base_inference_throughput_sync": 0.012165061711106812,
13
+ "base_inference_throughput_async": 0.012517552849551352,
14
  "smashed_current_gpu_type": "NVIDIA A100-PCIE-40GB",
15
+ "smashed_current_gpu_total_memory": 40339.3125,
16
+ "smashed_token_generation_latency_sync": 177.8312194824219,
17
+ "smashed_token_generation_latency_async": 178.8266021758318,
18
+ "smashed_token_generation_throughput_sync": 0.0056233095792206905,
19
+ "smashed_token_generation_throughput_async": 0.005592009174433382,
20
+ "smashed_token_generation_CO2_emissions": 5.6607409622391035e-05,
21
+ "smashed_token_generation_energy_consumption": 0.019573642151998485,
22
+ "smashed_inference_latency_sync": 185.82732849121095,
23
+ "smashed_inference_latency_async": 144.77639198303223,
24
+ "smashed_inference_throughput_sync": 0.0053813398067943325,
25
+ "smashed_inference_throughput_async": 0.0069072034901740045
26
  }
smash_config.json CHANGED
@@ -14,7 +14,7 @@
14
  "controlnet": "None",
15
  "unet_dim": 4,
16
  "device": "cuda",
17
- "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelst5pb_r7v",
18
  "batch_size": 1,
19
  "model_name": "mattshumer/Llama-3-8B-16K",
20
  "task": "text_text_generation",
 
14
  "controlnet": "None",
15
  "unet_dim": 4,
16
  "device": "cuda",
17
+ "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelstj7ks5m6",
18
  "batch_size": 1,
19
  "model_name": "mattshumer/Llama-3-8B-16K",
20
  "task": "text_text_generation",