cloudyu-nemo-dpo-v11

developer_uid: cloudyu

submission_id: cloudyu-nemo-dpo-v11_v1

model_name: cloudyu-nemo-dpo-v11_v1

model_group: cloudyu/Nemo-DPO-v11

status: inactive

timestamp: 2024-11-28T01:14:50+00:00

num_battles: 16172

num_wins: 7992

celo_rating: 1260.04

family_friendly_score: 0.5724

family_friendly_standard_error: 0.006996545433283485

submission_type: basic

model_repo: cloudyu/Nemo-DPO-v11

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.6246971801068393, 'latency_mean': 1.60070667386055, 'latency_p50': 1.5919040441513062, 'latency_p90': 1.7646998167037964}, {'batch_size': 3, 'throughput': 1.1335284865405335, 'latency_mean': 2.6419673919677735, 'latency_p50': 2.6610641479492188, 'latency_p90': 2.8515199422836304}, {'batch_size': 5, 'throughput': 1.3853598303249441, 'latency_mean': 3.5843556225299835, 'latency_p50': 3.5863471031188965, 'latency_p90': 3.990722823143005}, {'batch_size': 6, 'throughput': 1.4557620570345635, 'latency_mean': 4.100519889593125, 'latency_p50': 4.086998462677002, 'latency_p90': 4.622587180137634}, {'batch_size': 8, 'throughput': 1.514962995818895, 'latency_mean': 5.235363826751709, 'latency_p50': 5.279506087303162, 'latency_p90': 5.908110952377319}, {'batch_size': 10, 'throughput': 1.56960093601311, 'latency_mean': 6.324545018672943, 'latency_p50': 6.314915180206299, 'latency_p90': 7.1675190210342405}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: cloudyu-nemo-dpo-v11_v1

is_internal_developer: False

language_model: cloudyu/Nemo-DPO-v11

model_size: 13B

ranking_group: single

throughput_3p7s: 1.41

us_pacific_date: 2024-11-27

win_ratio: 0.4941874845411823

generation_params: {'temperature': 0.95, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}

Resubmit model

Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name cloudyu-nemo-dpo-v11-v1-mkmlizer
Waiting for job on cloudyu-nemo-dpo-v11-v1-mkmlizer to finish
cloudyu-nemo-dpo-v11-v1-mkmlizer: Downloaded to shared memory in 56.141s
cloudyu-nemo-dpo-v11-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmphplfsd6o, device:0
cloudyu-nemo-dpo-v11-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cloudyu-nemo-dpo-v11-v1-mkmlizer: quantized model in 36.802s
cloudyu-nemo-dpo-v11-v1-mkmlizer: Processed model cloudyu/Nemo-DPO-v11 in 92.944s
cloudyu-nemo-dpo-v11-v1-mkmlizer: creating bucket guanaco-mkml-models
cloudyu-nemo-dpo-v11-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cloudyu-nemo-dpo-v11-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1
cloudyu-nemo-dpo-v11-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1/config.json
cloudyu-nemo-dpo-v11-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1/special_tokens_map.json
cloudyu-nemo-dpo-v11-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1/tokenizer_config.json
cloudyu-nemo-dpo-v11-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1/tokenizer.json
cloudyu-nemo-dpo-v11-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cloudyu-nemo-dpo-v11-v1/flywheel_model.0.safetensors
cloudyu-nemo-dpo-v11-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/363 [00:06<18:36,  3.09s/it]
Loading 0:   2%|▏         | 6/363 [00:06<04:58,  1.20it/s]
Loading 0:   4%|▎         | 13/363 [00:06<01:46,  3.28it/s]
Loading 0:   5%|▍         | 17/363 [00:06<01:12,  4.80it/s]
Loading 0:   6%|▋         | 23/363 [00:06<00:43,  7.85it/s]
Loading 0:   8%|▊         | 29/363 [00:06<00:29, 11.27it/s]
Loading 0:   9%|▉         | 34/363 [00:07<00:22, 14.32it/s]
Loading 0:  11%|█         | 40/363 [00:07<00:19, 16.20it/s]
Loading 0:  12%|█▏        | 44/363 [00:07<00:17, 18.50it/s]
Loading 0:  14%|█▍        | 50/363 [00:07<00:13, 23.90it/s]
Loading 0:  15%|█▍        | 54/363 [00:07<00:11, 25.91it/s]
Loading 0:  16%|█▋        | 59/363 [00:07<00:10, 29.83it/s]
Loading 0:  18%|█▊        | 65/363 [00:07<00:09, 31.79it/s]
Loading 0:  19%|█▉        | 69/363 [00:08<00:09, 31.88it/s]
Loading 0:  21%|██        | 76/363 [00:08<00:07, 39.86it/s]
Loading 0:  22%|██▏       | 81/363 [00:08<00:06, 40.39it/s]
Loading 0:  24%|██▎       | 86/363 [00:08<00:06, 40.82it/s]
Loading 0:  25%|██▌       | 92/363 [00:08<00:06, 40.28it/s]
Loading 0:  27%|██▋       | 97/363 [00:08<00:06, 39.49it/s]
Loading 0:  28%|██▊       | 103/363 [00:08<00:05, 43.70it/s]
Loading 0:  30%|██▉       | 108/363 [00:08<00:05, 42.55it/s]
Loading 0:  31%|███       | 113/363 [00:09<00:05, 42.84it/s]
Loading 0:  33%|███▎      | 118/363 [00:09<00:05, 44.24it/s]
Loading 0:  34%|███▍      | 123/363 [00:09<00:09, 26.63it/s]
Loading 0:  36%|███▌      | 131/363 [00:09<00:06, 35.07it/s]
Loading 0:  37%|███▋      | 136/363 [00:09<00:05, 38.06it/s]
Loading 0:  39%|███▉      | 141/363 [00:09<00:06, 33.57it/s]
Loading 0:  41%|████      | 149/363 [00:10<00:05, 41.52it/s]
Loading 0:  42%|████▏     | 154/363 [00:10<00:04, 43.30it/s]
Loading 0:  44%|████▍     | 159/363 [00:10<00:05, 37.04it/s]
Loading 0:  46%|████▌     | 167/363 [00:10<00:04, 44.72it/s]
Loading 0:  48%|████▊     | 173/363 [00:10<00:04, 42.94it/s]
Loading 0:  49%|████▉     | 178/363 [00:10<00:04, 41.41it/s]
Loading 0:  51%|█████     | 184/363 [00:10<00:03, 45.67it/s]
Loading 0:  52%|█████▏    | 189/363 [00:10<00:03, 44.67it/s]
Loading 0:  53%|█████▎    | 194/363 [00:11<00:03, 45.11it/s]
Loading 0:  55%|█████▍    | 199/363 [00:11<00:03, 45.98it/s]
Loading 0:  56%|█████▌    | 204/363 [00:11<00:05, 26.63it/s]
Loading 0:  58%|█████▊    | 212/363 [00:11<00:04, 35.11it/s]
Loading 0:  60%|██████    | 218/363 [00:11<00:03, 36.74it/s]
Loading 0:  61%|██████▏   | 223/363 [00:11<00:03, 37.93it/s]
Loading 0:  63%|██████▎   | 229/363 [00:12<00:03, 40.91it/s]
Loading 0:  64%|██████▍   | 234/363 [00:12<00:03, 40.77it/s]
Loading 0:  66%|██████▌   | 239/363 [00:12<00:02, 41.50it/s]
Loading 0:  67%|██████▋   | 245/363 [00:12<00:02, 41.11it/s]
Loading 0:  69%|██████▉   | 250/363 [00:12<00:02, 41.27it/s]
Loading 0:  71%|███████   | 257/363 [00:12<00:02, 46.92it/s]
Loading 0:  72%|███████▏  | 263/363 [00:12<00:02, 45.55it/s]
Loading 0:  74%|███████▍  | 268/363 [00:12<00:02, 44.16it/s]
Loading 0:  76%|███████▌  | 275/363 [00:13<00:01, 49.17it/s]
Loading 0:  77%|███████▋  | 281/363 [00:13<00:01, 46.88it/s]
Loading 0:  79%|███████▉  | 286/363 [00:13<00:02, 31.16it/s]
Loading 0:  81%|████████  | 293/363 [00:13<00:01, 37.34it/s]
Loading 0:  82%|████████▏ | 299/363 [00:13<00:01, 38.49it/s]
Loading 0:  84%|████████▎ | 304/363 [00:13<00:01, 38.70it/s]
Loading 0:  85%|████████▌ | 310/363 [00:13<00:01, 42.53it/s]
Loading 0:  87%|████████▋ | 315/363 [00:14<00:01, 41.57it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:01, 41.90it/s]
Loading 0:  90%|████████▉ | 325/363 [00:14<00:00, 43.59it/s]
Loading 0:  91%|█████████ | 330/363 [00:14<00:00, 37.64it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:00, 45.76it/s]
Loading 0:  94%|█████████▍| 343/363 [00:14<00:00, 46.71it/s]
Loading 0:  96%|█████████▌| 348/363 [00:14<00:00, 38.65it/s]
Loading 0:  98%|█████████▊| 355/363 [00:15<00:00, 43.65it/s]
Loading 0:  99%|█████████▉| 360/363 [00:15<00:00, 43.60it/s]
                                                            
Job cloudyu-nemo-dpo-v11-v1-mkmlizer completed after 123.44s with status: succeeded
Stopping job with name cloudyu-nemo-dpo-v11-v1-mkmlizer
Pipeline stage MKMLizer completed in 123.95s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service cloudyu-nemo-dpo-v11-v1
Waiting for inference service cloudyu-nemo-dpo-v11-v1 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service cloudyu-nemo-dpo-v11-v1 ready after 130.46231341362s
Pipeline stage MKMLDeployer completed in 130.97s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0937390327453613s
Received healthy response to inference request in 1.633056402206421s
Received healthy response to inference request in 1.7209367752075195s
Received healthy response to inference request in 1.77921724319458s
Received healthy response to inference request in 1.7450175285339355s
5 requests
0 failed requests
5th percentile: 1.6506324768066407
10th percentile: 1.6682085514068603
20th percentile: 1.7033607006072997
30th percentile: 1.7257529258728028
40th percentile: 1.7353852272033692
50th percentile: 1.7450175285339355
60th percentile: 1.7586974143981933
70th percentile: 1.7723773002624512
80th percentile: 1.8421216011047363
90th percentile: 1.9679303169250488
95th percentile: 2.030834674835205
99th percentile: 2.08115816116333
mean time: 1.7943933963775636
Pipeline stage StressChecker completed in 10.25s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 2.23s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 2.08s
Shutdown handler de-registered
cloudyu-nemo-dpo-v11_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 3358.48s
Shutdown handler de-registered
cloudyu-nemo-dpo-v11_v1 status is now inactive due to auto deactivation removed underperforming models