rirv938-anthropic-grpo-4_3554

developer_uid: rirv938

submission_id: rirv938-anthropic-grpo-4_3554_v1

model_name: rirv938-anthropic-grpo-4_3554_v1

model_group: rirv938/anthropic_grpo_4

status: torndown

timestamp: 2025-04-30T15:00:50+00:00

num_battles: 9025

num_wins: 4968

celo_rating: 1340.76

family_friendly_score: 0.526

family_friendly_standard_error: 0.007061501256815012

submission_type: basic

model_repo: rirv938/anthropic_grpo_40k_cp888_0ff_merged

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 768

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6453296552844993, 'latency_mean': 1.549489461183548, 'latency_p50': 1.5398987531661987, 'latency_p90': 1.7168253898620605}, {'batch_size': 3, 'throughput': 1.228463304481041, 'latency_mean': 2.4315641367435457, 'latency_p50': 2.434945583343506, 'latency_p90': 2.669797587394714}, {'batch_size': 5, 'throughput': 1.5204977918617886, 'latency_mean': 3.2774340319633484, 'latency_p50': 3.2777403593063354, 'latency_p90': 3.613864731788635}, {'batch_size': 6, 'throughput': 1.6092340688688795, 'latency_mean': 3.706691651344299, 'latency_p50': 3.7176088094711304, 'latency_p90': 4.126570200920105}, {'batch_size': 8, 'throughput': 1.6988624722522787, 'latency_mean': 4.685649515390396, 'latency_p50': 4.687879920005798, 'latency_p90': 5.17727313041687}, {'batch_size': 10, 'throughput': 1.749270715103943, 'latency_mean': 5.659246082305908, 'latency_p50': 5.665336966514587, 'latency_p90': 6.35122663974762}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rirv938-anthropic-grpo-4_3554_v1

is_internal_developer: True

language_model: rirv938/anthropic_grpo_40k_cp888_0ff_merged

model_size: 13B

ranking_group: single

throughput_3p7s: 1.62

us_pacific_date: 2025-04-30

win_ratio: 0.5504709141274238

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '###', '</s>', 'You:'], 'max_input_tokens': 768, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '', 'prompt_template': '', '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 rirv938-anthropic-grpo-4-3554-v1-mkmlizer
Waiting for job on rirv938-anthropic-grpo-4-3554-v1-mkmlizer to finish
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║     _____            __           __                                ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║       /___/                                                         ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║                                                                     ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║                                                                     ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  belonging to:                                                      ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║                                                                     ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ║                                                                     ║
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: Downloaded to shared memory in 92.508s
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmphr524fep, device:0
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: quantized model in 42.549s
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: Processed model rirv938/anthropic_grpo_40k_cp888_0ff_merged in 135.057s
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1/config.json
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1/special_tokens_map.json
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1/tokenizer_config.json
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1/tokenizer.json
rirv938-anthropic-grpo-4-3554-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-anthropic-grpo-4-3554-v1/flywheel_model.0.safetensors
Job rirv938-anthropic-grpo-4-3554-v1-mkmlizer completed after 166.56s with status: succeeded
Stopping job with name rirv938-anthropic-grpo-4-3554-v1-mkmlizer
Pipeline stage MKMLizer completed in 167.03s
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-anthropic-grpo-4-3554-v1
Waiting for inference service rirv938-anthropic-grpo-4-3554-v1 to be ready
Failed to get response for submission blend_kinam_2025-04-14: ('http://rirv938-anthropic-40k-2-43413-v8-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:46538->127.0.0.1:8080: read: connection reset by peer\n')
Inference service rirv938-anthropic-grpo-4-3554-v1 ready after 141.02373003959656s
Pipeline stage MKMLDeployer completed in 141.46s
run pipeline stage %s
Running pipeline stage StressChecker
HTTPConnectionPool(host='guanaco-submitter.guanaco-backend.k2.chaiverse.com', port=80): Read timed out. (read timeout=20)
Received unhealthy response to inference request!
Received healthy response to inference request in 2.082794427871704s
Received healthy response to inference request in 1.9080932140350342s
Received healthy response to inference request in 1.7630386352539062s
Received healthy response to inference request in 1.5691602230072021s
5 requests
1 failed requests
5th percentile: 1.607935905456543
10th percentile: 1.6467115879058838
20th percentile: 1.7242629528045654
30th percentile: 1.7920495510101317
40th percentile: 1.850071382522583
50th percentile: 1.9080932140350342
60th percentile: 1.9779736995697021
70th percentile: 2.04785418510437
80th percentile: 5.693691635131839
90th percentile: 12.915486049652102
95th percentile: 16.52638325691223
99th percentile: 19.415101022720336
mean time: 5.492073392868042
%s, retrying in %s seconds...
Received healthy response to inference request in 1.6416468620300293s
Received healthy response to inference request in 2.158872604370117s
Received healthy response to inference request in 1.8194289207458496s
Received healthy response to inference request in 1.9060778617858887s
Received healthy response to inference request in 6.714556455612183s
5 requests
0 failed requests
5th percentile: 1.6772032737731934
10th percentile: 1.7127596855163574
20th percentile: 1.7838725090026855
30th percentile: 1.8367587089538575
40th percentile: 1.871418285369873
50th percentile: 1.9060778617858887
60th percentile: 2.00719575881958
70th percentile: 2.1083136558532716
80th percentile: 3.070009374618531
90th percentile: 4.892282915115357
95th percentile: 5.803419685363769
99th percentile: 6.5323291015625
mean time: 2.8481165409088134
Pipeline stage StressChecker completed in 44.37s
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 0.73s
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 0.79s
Shutdown handler de-registered
rirv938-anthropic-grpo-4_3554_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 3161.72s
Shutdown handler de-registered
rirv938-anthropic-grpo-4_3554_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-anthropic-grpo-4_3554_v1 status is now torndown due to DeploymentManager action