developer_uid: rirv938
submission_id: rirv938-grpo-20250711-c_63518_v2
model_name: rirv938-grpo-20250711-c_63518_v2
model_group: rirv938/grpo_20250711_cp
status: torndown
timestamp: 2025-07-12T04:26:22+00:00
num_battles: 8605
num_wins: 4395
celo_rating: 1294.86
family_friendly_score: 0.477
family_friendly_standard_error: 0.007063582660378514
submission_type: basic
model_repo: rirv938/grpo_20250711_cp312_sid_mistral_24b_dpo_40k__merged
model_architecture: MistralForCausalLM
model_num_parameters: 24096691200.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.5171216627115337, 'latency_mean': 1.9336565244197845, 'latency_p50': 1.9415109157562256, 'latency_p90': 2.139647054672241}, {'batch_size': 3, 'throughput': 1.0592700809002704, 'latency_mean': 2.8202871334552766, 'latency_p50': 2.808563470840454, 'latency_p90': 3.1449411869049073}, {'batch_size': 5, 'throughput': 1.3541119504167247, 'latency_mean': 3.6759383809566497, 'latency_p50': 3.6884660720825195, 'latency_p90': 4.085365986824035}, {'batch_size': 6, 'throughput': 1.4571906371037897, 'latency_mean': 4.091835905313491, 'latency_p50': 4.0861334800720215, 'latency_p90': 4.5422121524810795}, {'batch_size': 8, 'throughput': 1.5838190586535863, 'latency_mean': 5.014606553316116, 'latency_p50': 5.059622645378113, 'latency_p90': 5.652696704864502}, {'batch_size': 10, 'throughput': 1.6681569717787608, 'latency_mean': 5.925662469863892, 'latency_p50': 5.937679052352905, 'latency_p90': 6.651551914215088}]
gpu_counts: {'NVIDIA A100-SXM4-80GB': 1}
display_name: rirv938-grpo-20250711-c_63518_v2
is_internal_developer: True
language_model: rirv938/grpo_20250711_cp312_sid_mistral_24b_dpo_40k__merged
model_size: 24B
ranking_group: single
throughput_3p7s: 1.37
us_pacific_date: 2025-07-11
win_ratio: 0.5107495642068565
generation_params: {'temperature': 0.9, 'top_p': 0.9, 'min_p': 0.6, 'top_k': 80, 'presence_penalty': 0.5, 'frequency_penalty': 0.5, 'stopping_words': ['You:', '###', '\n', '</s>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': '[INST]', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '[/INST]{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 rirv938-grpo-20250711-c-63518-v2-mkmlizer
Waiting for job on rirv938-grpo-20250711-c-63518-v2-mkmlizer to finish
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Version: 0.29.15                                                   ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  belonging to:                                                      ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ║                                                                     ║
rirv938-grpo-20250711-c-63518-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Downloaded to shared memory in 128.151s
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Checking if rirv938/grpo_20250711_cp312_sid_mistral_24b_dpo_40k__merged already exists in ChaiML
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Creating repo ChaiML/grpo_20250711_cp312_sid_mistral_24b_dpo_40k__merged and uploading /tmp/tmpf8ifo97i to it
rirv938-grpo-20250711-c-63518-v2-mkmlizer: 
  0%|          | 0/22 [00:00<?, ?it/s]
  5%|▍         | 1/22 [00:03<01:20,  3.84s/it]
  9%|▉         | 2/22 [00:07<01:13,  3.70s/it]
 14%|█▎        | 3/22 [00:11<01:09,  3.67s/it]
 18%|█▊        | 4/22 [00:18<01:29,  4.96s/it]
 23%|██▎       | 5/22 [00:26<01:44,  6.14s/it]
 27%|██▋       | 6/22 [00:30<01:27,  5.45s/it]
 32%|███▏      | 7/22 [00:38<01:35,  6.35s/it]
 36%|███▋      | 8/22 [00:46<01:35,  6.85s/it]
 41%|████      | 9/22 [00:50<01:15,  5.84s/it]
 45%|████▌     | 10/22 [00:54<01:05,  5.47s/it]
 50%|█████     | 11/22 [00:58<00:53,  4.84s/it]
 55%|█████▍    | 12/22 [01:01<00:44,  4.42s/it]
 59%|█████▉    | 13/22 [01:05<00:37,  4.11s/it]
 64%|██████▎   | 14/22 [01:09<00:33,  4.21s/it]
 68%|██████▊   | 15/22 [01:13<00:28,  4.11s/it]
 73%|███████▎  | 16/22 [01:20<00:30,  5.13s/it]
 77%|███████▋  | 17/22 [01:24<00:23,  4.63s/it]
 82%|████████▏ | 18/22 [01:27<00:17,  4.34s/it]
 86%|████████▋ | 19/22 [01:32<00:12,  4.33s/it]
 91%|█████████ | 20/22 [01:35<00:08,  4.07s/it]
 95%|█████████▌| 21/22 [02:11<00:13, 13.64s/it]
100%|██████████| 22/22 [02:12<00:00,  9.88s/it]
100%|██████████| 22/22 [02:12<00:00,  6.04s/it]
rirv938-grpo-20250711-c-63518-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpf8ifo97i, device:0
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-grpo-20250711-c-63518-v2-mkmlizer: quantized model in 55.245s
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Processed model rirv938/grpo_20250711_cp312_sid_mistral_24b_dpo_40k__merged in 407.552s
rirv938-grpo-20250711-c-63518-v2-mkmlizer: creating bucket guanaco-mkml-models
rirv938-grpo-20250711-c-63518-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-grpo-20250711-c-63518-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/config.json
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/special_tokens_map.json
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/tokenizer_config.json
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/tokenizer.json
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/flywheel_model.0.safetensors
rirv938-grpo-20250711-c-63518-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-grpo-20250711-c-63518-v2/nvidia/flywheel_model.1.safetensors
Job rirv938-grpo-20250711-c-63518-v2-mkmlizer completed after 443.88s with status: succeeded
Stopping job with name rirv938-grpo-20250711-c-63518-v2-mkmlizer
Pipeline stage MKMLizer completed in 444.37s
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 rirv938-grpo-20250711-c-63518-v2
Waiting for inference service rirv938-grpo-20250711-c-63518-v2 to be ready
Inference service rirv938-grpo-20250711-c-63518-v2 ready after 201.51086020469666s
Pipeline stage MKMLDeployer completed in 202.07s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4443984031677246s
Received healthy response to inference request in 2.1291966438293457s
Received healthy response to inference request in 1.9878182411193848s
Received healthy response to inference request in 2.1229560375213623s
Received healthy response to inference request in 1.8056590557098389s
5 requests
0 failed requests
5th percentile: 1.842090892791748
10th percentile: 1.8785227298736573
20th percentile: 1.9513864040374755
30th percentile: 2.0148458003997805
40th percentile: 2.0689009189605714
50th percentile: 2.1229560375213623
60th percentile: 2.1254522800445557
70th percentile: 2.127948522567749
80th percentile: 2.1922369956970216
90th percentile: 2.318317699432373
95th percentile: 2.3813580513000487
99th percentile: 2.4317903327941894
mean time: 2.098005676269531
Pipeline stage StressChecker completed in 11.74s
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 0.74s
Shutdown handler de-registered
rirv938-grpo-20250711-c_63518_v2 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 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-grpo-20250711-c-63518-v2-profiler
Waiting for inference service rirv938-grpo-20250711-c-63518-v2-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 3190.58s
Shutdown handler de-registered
rirv938-grpo-20250711-c_63518_v2 status is now inactive due to auto deactivation removed underperforming models
rirv938-grpo-20250711-c_63518_v2 status is now torndown due to DeploymentManager action