developer_uid: AlexCuadron
submission_id: alexcuadron-dpo-slurp-r_29528_v1
model_name: alexcuadron-dpo-slurp-r_29528_v1
model_group: AlexCuadron/dpo_slurp_ro
status: torndown
timestamp: 2025-03-24T19:04:16+00:00
num_battles: 7180
num_wins: 3626
celo_rating: 1279.26
family_friendly_score: 0.5744
family_friendly_standard_error: 0.006992347817435857
submission_type: basic
model_repo: AlexCuadron/dpo_slurp_roleplay_v2
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.5952705066060618, 'latency_mean': 1.6798235487937927, 'latency_p50': 1.6801069974899292, 'latency_p90': 1.847358274459839}, {'batch_size': 3, 'throughput': 1.0914362300730962, 'latency_mean': 2.735801384449005, 'latency_p50': 2.738491177558899, 'latency_p90': 2.9936434030532837}, {'batch_size': 5, 'throughput': 1.3060014616129925, 'latency_mean': 3.8059656500816343, 'latency_p50': 3.7952502965927124, 'latency_p90': 4.1987015962600704}, {'batch_size': 6, 'throughput': 1.3723804286552543, 'latency_mean': 4.356865776777267, 'latency_p50': 4.372917413711548, 'latency_p90': 4.858129525184631}, {'batch_size': 8, 'throughput': 1.4481639582994434, 'latency_mean': 5.501195833683014, 'latency_p50': 5.529194355010986, 'latency_p90': 6.064143180847168}, {'batch_size': 10, 'throughput': 1.472553911907814, 'latency_mean': 6.736948103904724, 'latency_p50': 6.679298520088196, 'latency_p90': 7.678029012680054}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: alexcuadron-dpo-slurp-r_29528_v1
is_internal_developer: False
language_model: AlexCuadron/dpo_slurp_roleplay_v2
model_size: 13B
ranking_group: single
throughput_3p7s: 1.3
us_pacific_date: 2025-03-24
win_ratio: 0.5050139275766017
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.1, 'frequency_penalty': 0.0, 'stopping_words': ['</s>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': "<s>{bot_name}'s Persona: {memory}\n", 'prompt_template': '{prompt}', 'bot_template': '{bot_name}: {message}</s>', 'user_template': '[INST]{user_name}: {message}[/INST]', '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 alexcuadron-dpo-slurp-r-29528-v1-mkmlizer
Waiting for job on alexcuadron-dpo-slurp-r-29528-v1-mkmlizer to finish
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ _____ __ __ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ /___/ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ Version: 0.12.8 ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ https://mk1.ai ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ The license key for the current software has been verified as ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ belonging to: ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ Chai Research Corp. ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ Expiration: 2025-04-15 23:59:59 ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ║ ║
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: Downloaded to shared memory in 43.229s
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpn5wj7trx, device:0
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: quantized model in 38.634s
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: Processed model AlexCuadron/dpo_slurp_roleplay_v2 in 81.863s
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: creating bucket guanaco-mkml-models
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/alexcuadron-dpo-slurp-r-29528-v1
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/alexcuadron-dpo-slurp-r-29528-v1/config.json
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/alexcuadron-dpo-slurp-r-29528-v1/tokenizer_config.json
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/alexcuadron-dpo-slurp-r-29528-v1/tokenizer.json
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/alexcuadron-dpo-slurp-r-29528-v1/special_tokens_map.json
alexcuadron-dpo-slurp-r-29528-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%| | 2/363 [00:06<18:43, 3.11s/it] Loading 0: 2%|▏ | 6/363 [00:06<04:59, 1.19it/s] Loading 0: 3%|▎ | 11/363 [00:06<02:11, 2.67it/s] Loading 0: 4%|▍ | 15/363 [00:06<01:24, 4.14it/s] Loading 0: 6%|▌ | 22/363 [00:06<00:44, 7.71it/s] Loading 0: 7%|▋ | 26/363 [00:06<00:34, 9.84it/s] Loading 0: 9%|▊ | 31/363 [00:07<00:24, 13.31it/s] Loading 0: 10%|▉ | 35/363 [00:07<00:20, 15.80it/s] Loading 0: 11%|█ | 40/363 [00:07<00:20, 15.68it/s] Loading 0: 12%|█▏ | 43/363 [00:07<00:19, 16.77it/s] Loading 0: 13%|█▎ | 47/363 [00:07<00:15, 20.12it/s] Loading 0: 14%|█▍ | 51/363 [00:07<00:14, 21.28it/s] Loading 0: 15%|█▌ | 56/363 [00:07<00:11, 26.21it/s] Loading 0: 17%|█▋ | 60/363 [00:08<00:11, 26.15it/s] Loading 0: 18%|█▊ | 65/363 [00:08<00:09, 30.99it/s] Loading 0: 19%|█▉ | 69/363 [00:08<00:09, 29.62it/s] Loading 0: 20%|██ | 74/363 [00:08<00:08, 34.16it/s] Loading 0: 21%|██▏ | 78/363 [00:08<00:09, 31.50it/s] Loading 0: 23%|██▎ | 83/363 [00:08<00:07, 35.43it/s] Loading 0: 24%|██▍ | 87/363 [00:08<00:08, 32.17it/s] Loading 0: 25%|██▌ | 92/363 [00:09<00:07, 35.77it/s] Loading 0: 26%|██▋ | 96/363 [00:09<00:08, 32.14it/s] Loading 0: 28%|██▊ | 101/363 [00:09<00:07, 35.36it/s] Loading 0: 29%|██▉ | 105/363 [00:09<00:08, 31.94it/s] Loading 0: 30%|███ | 110/363 [00:09<00:07, 35.30it/s] Loading 0: 31%|███▏ | 114/363 [00:09<00:07, 31.22it/s] Loading 0: 33%|███▎ | 119/363 [00:09<00:06, 35.00it/s] Loading 0: 34%|███▍ | 123/363 [00:10<00:10, 23.82it/s] Loading 0: 36%|███▌ | 130/363 [00:10<00:07, 31.27it/s] Loading 0: 37%|███▋ | 134/363 [00:10<00:07, 32.34it/s] Loading 0: 39%|███▊ | 140/363 [00:10<00:05, 37.45it/s] Loading 0: 40%|████ | 146/363 [00:10<00:05, 38.71it/s] Loading 0: 42%|████▏ | 151/363 [00:10<00:05, 38.76it/s] Loading 0: 43%|████▎ | 157/363 [00:10<00:04, 42.94it/s] Loading 0: 45%|████▍ | 162/363 [00:10<00:04, 43.67it/s] Loading 0: 46%|████▌ | 167/363 [00:11<00:04, 44.55it/s] Loading 0: 48%|████▊ | 173/363 [00:11<00:04, 42.87it/s] Loading 0: 49%|████▉ | 178/363 [00:11<00:04, 41.62it/s] Loading 0: 51%|█████ | 185/363 [00:11<00:03, 46.40it/s] Loading 0: 53%|█████▎ | 191/363 [00:11<00:04, 42.81it/s] Loading 0: 54%|█████▍ | 196/363 [00:11<00:04, 38.65it/s] Loading 0: 55%|█████▌ | 200/363 [00:11<00:04, 38.52it/s] Loading 0: 56%|█████▌ | 204/363 [00:12<00:06, 23.89it/s] Loading 0: 58%|█████▊ | 209/363 [00:12<00:05, 28.00it/s] Loading 0: 59%|█████▊ | 213/363 [00:12<00:05, 27.19it/s] Loading 0: 60%|██████ | 218/363 [00:12<00:04, 31.11it/s] Loading 0: 61%|██████ | 222/363 [00:12<00:04, 28.76it/s] Loading 0: 63%|██████▎ | 227/363 [00:12<00:04, 32.90it/s] Loading 0: 64%|██████▎ | 231/363 [00:13<00:04, 30.39it/s] Loading 0: 65%|██████▌ | 236/363 [00:13<00:03, 34.33it/s] Loading 0: 66%|██████▌ | 240/363 [00:13<00:04, 30.46it/s] Loading 0: 67%|██████▋ | 245/363 [00:13<00:03, 33.85it/s] Loading 0: 69%|██████▊ | 249/363 [00:13<00:03, 30.70it/s] Loading 0: 70%|██████▉ | 254/363 [00:13<00:03, 34.30it/s] Loading 0: 71%|███████ | 258/363 [00:13<00:03, 30.51it/s] Loading 0: 72%|███████▏ | 263/363 [00:14<00:02, 34.13it/s] Loading 0: 74%|███████▎ | 267/363 [00:14<00:03, 30.70it/s] Loading 0: 75%|███████▍ | 272/363 [00:14<00:02, 33.99it/s] Loading 0: 76%|███████▌ | 276/363 [00:14<00:02, 29.77it/s] Loading 0: 77%|███████▋ | 281/363 [00:14<00:02, 33.78it/s] Loading 0: 79%|███████▊ | 285/363 [00:14<00:03, 20.68it/s] Loading 0: 80%|███████▉ | 290/363 [00:15<00:02, 25.16it/s] Loading 0: 81%|████████ | 294/363 [00:15<00:02, 24.79it/s] Loading 0: 82%|████████▏ | 299/363 [00:15<00:02, 28.69it/s] Loading 0: 83%|████████▎ | 303/363 [00:15<00:02, 27.62it/s] Loading 0: 85%|████████▍ | 308/363 [00:15<00:01, 32.00it/s] Loading 0: 86%|████████▌ | 312/363 [00:15<00:01, 29.91it/s] Loading 0: 87%|████████▋ | 317/363 [00:15<00:01, 34.37it/s] Loading 0: 88%|████████▊ | 321/363 [00:16<00:01, 32.56it/s] Loading 0: 90%|█████████ | 328/363 [00:16<00:00, 40.76it/s] Loading 0: 92%|█████████▏| 333/363 [00:16<00:00, 41.47it/s] Loading 0: 93%|█████████▎| 338/363 [00:16<00:00, 41.92it/s] Loading 0: 95%|█████████▍| 344/363 [00:16<00:00, 42.32it/s] Loading 0: 96%|█████████▌| 349/363 [00:16<00:00, 41.81it/s] Loading 0: 98%|█████████▊| 355/363 [00:16<00:00, 46.31it/s] Loading 0: 99%|█████████▉| 360/363 [00:16<00:00, 45.81it/s]
Job alexcuadron-dpo-slurp-r-29528-v1-mkmlizer completed after 117.8s with status: succeeded
Stopping job with name alexcuadron-dpo-slurp-r-29528-v1-mkmlizer
Pipeline stage MKMLizer completed in 118.28s
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 alexcuadron-dpo-slurp-r-29528-v1
Waiting for inference service alexcuadron-dpo-slurp-r-29528-v1 to be ready
Inference service alexcuadron-dpo-slurp-r-29528-v1 ready after 90.33882164955139s
Pipeline stage MKMLDeployer completed in 90.84s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2575690746307373s
Received healthy response to inference request in 1.8200726509094238s
Received healthy response to inference request in 1.785649299621582s
Received healthy response to inference request in 1.9116594791412354s
Received healthy response to inference request in 1.7993123531341553s
5 requests
0 failed requests
5th percentile: 1.7883819103240968
10th percentile: 1.7911145210266113
20th percentile: 1.7965797424316405
30th percentile: 1.803464412689209
40th percentile: 1.8117685317993164
50th percentile: 1.8200726509094238
60th percentile: 1.8567073822021485
70th percentile: 1.8933421134948731
80th percentile: 1.9808413982391357
90th percentile: 2.1192052364349365
95th percentile: 2.188387155532837
99th percentile: 2.2437326908111572
mean time: 1.9148525714874267
Pipeline stage StressChecker completed in 10.83s
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 0.71s
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 0.63s
Shutdown handler de-registered
alexcuadron-dpo-slurp-r_29528_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service alexcuadron-dpo-slurp-r-29528-v1-profiler
Waiting for inference service alexcuadron-dpo-slurp-r-29528-v1-profiler to be ready
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 2802.42s
Shutdown handler de-registered
alexcuadron-dpo-slurp-r_29528_v1 status is now inactive due to auto deactivation removed underperforming models
alexcuadron-dpo-slurp-r_29528_v1 status is now torndown due to DeploymentManager action
ChatRequest
Generation Params
Prompt Formatter
Chat History
ChatMessage 1