developer_uid: azuruce
submission_id: yamatazen-ayla-light-16b_v1
model_name: yamatazen-ayla-light-16b_v1
model_group: yamatazen/Ayla-Light-16B
status: torndown
timestamp: 2025-02-27T00:45:39+00:00
num_battles: 5611
num_wins: 2670
celo_rating: 1239.92
family_friendly_score: 0.5568
family_friendly_standard_error: 0.007025293730514049
submission_type: basic
model_repo: yamatazen/Ayla-Light-16B
model_architecture: MistralForCausalLM
model_num_parameters: 16772531200.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.46846689462861524, 'latency_mean': 2.1345608544349672, 'latency_p50': 2.1438708305358887, 'latency_p90': 2.3570125102996826}, {'batch_size': 3, 'throughput': 0.8663340351707075, 'latency_mean': 3.450492070913315, 'latency_p50': 3.4523251056671143, 'latency_p90': 3.763848304748535}, {'batch_size': 5, 'throughput': 1.050061757319398, 'latency_mean': 4.7372004091739655, 'latency_p50': 4.695827841758728, 'latency_p90': 5.362735676765442}, {'batch_size': 6, 'throughput': 1.0983190323500918, 'latency_mean': 5.440462152957917, 'latency_p50': 5.408968091011047, 'latency_p90': 6.051852178573609}, {'batch_size': 10, 'throughput': 1.182203456533221, 'latency_mean': 8.389196841716766, 'latency_p50': 8.431113004684448, 'latency_p90': 9.520441436767578}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: yamatazen-ayla-light-16b_v1
is_internal_developer: False
language_model: yamatazen/Ayla-Light-16B
model_size: 17B
ranking_group: single
throughput_3p7s: 0.92
us_pacific_date: 2025-02-26
win_ratio: 0.4758510069506327
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>', '</s>', 'Bot:', 'User:', 'Me:', 'You:', '###', '<|end_of_text|>', '####', '<|im_end|>'], 'max_input_tokens': 1024, '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 yamatazen-ayla-light-16b-v1-mkmlizer
Waiting for job on yamatazen-ayla-light-16b-v1-mkmlizer to finish
yamatazen-ayla-light-16b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
yamatazen-ayla-light-16b-v1-mkmlizer: ║ _____ __ __ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ /___/ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ Version: 0.12.8 ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ https://mk1.ai ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ belonging to: ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ Chai Research Corp. ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ Expiration: 2025-04-15 23:59:59 ║
yamatazen-ayla-light-16b-v1-mkmlizer: ║ ║
yamatazen-ayla-light-16b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
yamatazen-ayla-light-16b-v1-mkmlizer: Downloaded to shared memory in 60.171s
yamatazen-ayla-light-16b-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp6ueetec5, device:0
yamatazen-ayla-light-16b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
yamatazen-ayla-light-16b-v1-mkmlizer: quantized model in 45.852s
yamatazen-ayla-light-16b-v1-mkmlizer: Processed model yamatazen/Ayla-Light-16B in 106.024s
yamatazen-ayla-light-16b-v1-mkmlizer: creating bucket guanaco-mkml-models
yamatazen-ayla-light-16b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
yamatazen-ayla-light-16b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/special_tokens_map.json
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/config.json
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/tokenizer_config.json
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/tokenizer.json
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/flywheel_model.1.safetensors
yamatazen-ayla-light-16b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/yamatazen-ayla-light-16b-v1/flywheel_model.0.safetensors
yamatazen-ayla-light-16b-v1-mkmlizer: Loading 0: 0%| | 0/489 [00:00<?, ?it/s] Loading 0: 0%| | 2/489 [00:06<25:17, 3.12s/it] Loading 0: 1%| | 6/489 [00:06<06:46, 1.19it/s] Loading 0: 2%|▏ | 11/489 [00:06<02:59, 2.67it/s] Loading 0: 3%|▎ | 15/489 [00:06<01:53, 4.17it/s] Loading 0: 4%|▍ | 22/489 [00:06<00:59, 7.82it/s] Loading 0: 6%|▌ | 27/489 [00:06<00:43, 10.72it/s] Loading 0: 7%|▋ | 32/489 [00:07<00:32, 14.02it/s] Loading 0: 8%|▊ | 37/489 [00:07<00:25, 17.99it/s] Loading 0: 9%|▊ | 42/489 [00:07<00:28, 15.63it/s] Loading 0: 10%|█ | 49/489 [00:07<00:20, 21.55it/s] Loading 0: 11%|█ | 54/489 [00:07<00:17, 24.72it/s] Loading 0: 12%|█▏ | 58/489 [00:07<00:15, 26.96it/s] Loading 0: 13%|█▎ | 62/489 [00:08<00:14, 28.48it/s] Loading 0: 14%|█▎ | 67/489 [00:08<00:13, 32.20it/s] Loading 0: 15%|█▍ | 71/489 [00:08<00:12, 32.16it/s] Loading 0: 16%|█▌ | 76/489 [00:08<00:12, 33.89it/s] Loading 0: 16%|█▋ | 80/489 [00:08<00:12, 33.63it/s] Loading 0: 17%|█▋ | 85/489 [00:08<00:11, 36.50it/s] Loading 0: 18%|█▊ | 90/489 [00:08<00:11, 34.41it/s] Loading 0: 20%|█▉ | 97/489 [00:08<00:09, 41.36it/s] Loading 0: 21%|██ | 102/489 [00:09<00:10, 37.07it/s] Loading 0: 22%|██▏ | 106/489 [00:09<00:10, 36.80it/s] Loading 0: 22%|██▏ | 110/489 [00:09<00:10, 35.77it/s] Loading 0: 23%|██▎ | 114/489 [00:09<00:18, 20.27it/s] Loading 0: 25%|██▍ | 120/489 [00:09<00:14, 25.75it/s] Loading 0: 25%|██▌ | 124/489 [00:09<00:13, 27.27it/s] Loading 0: 26%|██▋ | 129/489 [00:10<00:11, 30.26it/s] Loading 0: 27%|██▋ | 133/489 [00:10<00:11, 30.85it/s] Loading 0: 28%|██▊ | 138/489 [00:10<00:10, 33.99it/s] Loading 0: 29%|██▉ | 142/489 [00:10<00:10, 33.98it/s] Loading 0: 30%|███ | 148/489 [00:10<00:08, 40.07it/s] Loading 0: 31%|███▏ | 153/489 [00:10<00:08, 38.15it/s] Loading 0: 32%|███▏ | 158/489 [00:10<00:08, 39.33it/s] Loading 0: 33%|███▎ | 163/489 [00:11<00:10, 32.15it/s] Loading 0: 35%|███▍ | 170/489 [00:11<00:08, 39.23it/s] Loading 0: 36%|███▌ | 176/489 [00:11<00:07, 43.80it/s] Loading 0: 37%|███▋ | 181/489 [00:11<00:07, 41.41it/s] Loading 0: 38%|███▊ | 187/489 [00:11<00:06, 44.23it/s] Loading 0: 39%|███▉ | 192/489 [00:11<00:06, 43.53it/s] Loading 0: 40%|████ | 197/489 [00:11<00:06, 43.10it/s] Loading 0: 41%|████▏ | 202/489 [00:12<00:09, 28.98it/s] Loading 0: 42%|████▏ | 206/489 [00:12<00:09, 30.21it/s] Loading 0: 43%|████▎ | 211/489 [00:12<00:08, 34.23it/s] Loading 0: 44%|████▍ | 215/489 [00:12<00:07, 34.68it/s] Loading 0: 45%|████▍ | 220/489 [00:12<00:07, 38.11it/s] Loading 0: 46%|████▌ | 225/489 [00:12<00:08, 32.31it/s] Loading 0: 47%|████▋ | 231/489 [00:12<00:07, 32.92it/s] Loading 0: 49%|████▊ | 238/489 [00:12<00:06, 39.44it/s] Loading 0: 50%|████▉ | 243/489 [00:13<00:06, 39.66it/s] Loading 0: 51%|█████ | 248/489 [00:13<00:05, 40.28it/s] Loading 0: 52%|█████▏ | 253/489 [00:13<00:05, 41.95it/s] Loading 0: 53%|█████▎ | 258/489 [00:13<00:06, 35.16it/s] Loading 0: 54%|█████▍ | 263/489 [00:13<00:05, 37.90it/s] Loading 0: 55%|█████▍ | 268/489 [00:13<00:05, 36.93it/s] Loading 0: 56%|█████▌ | 274/489 [00:13<00:05, 40.02it/s] Loading 0: 57%|█████▋ | 279/489 [00:14<00:05, 40.01it/s] Loading 0: 58%|█████▊ | 284/489 [00:14<00:06, 29.44it/s] Loading 0: 59%|█████▉ | 288/489 [00:14<00:06, 29.68it/s] Loading 0: 60%|█████▉ | 292/489 [00:14<00:06, 30.94it/s] Loading 0: 61%|██████ | 296/489 [00:14<00:06, 32.10it/s] Loading 0: 62%|██████▏ | 301/489 [00:14<00:05, 35.69it/s] Loading 0: 62%|██████▏ | 305/489 [00:14<00:05, 35.21it/s] Loading 0: 63%|██████▎ | 310/489 [00:15<00:04, 37.72it/s] Loading 0: 64%|██████▍ | 314/489 [00:15<00:04, 36.57it/s] Loading 0: 65%|██████▌ | 319/489 [00:15<00:04, 38.12it/s] Loading 0: 66%|██████▌ | 323/489 [00:15<00:04, 37.03it/s] Loading 0: 67%|██████▋ | 328/489 [00:15<00:04, 39.90it/s] Loading 0: 68%|██████▊ | 333/489 [00:15<00:03, 40.61it/s] Loading 0: 69%|██████▉ | 338/489 [00:15<00:03, 40.18it/s] Loading 0: 70%|███████ | 343/489 [00:15<00:03, 41.96it/s] Loading 0: 71%|███████ | 348/489 [00:16<00:04, 33.92it/s] Loading 0: 73%|███████▎ | 355/489 [00:16<00:03, 40.99it/s] Loading 0: 74%|███████▎ | 360/489 [00:16<00:03, 40.23it/s] Loading 0: 75%|███████▍ | 365/489 [00:16<00:04, 28.25it/s] Loading 0: 75%|███████▌ | 369/489 [00:16<00:04, 29.53it/s] Loading 0: 76%|███████▋ | 373/489 [00:16<00:03, 30.52it/s] Loading 0: 77%|███████▋ | 377/489 [00:16<00:03, 31.05it/s] Loading 0: 78%|███████▊ | 381/489 [00:31<00:03, 31.05it/s] Loading 0: 78%|███████▊ | 382/489 [00:31<01:43, 1.04it/s] Loading 0: 79%|███████▊ | 384/489 [00:31<01:26, 1.22it/s] Loading 0: 80%|███████▉ | 389/489 [00:31<00:53, 1.86it/s] Loading 0: 80%|████████ | 393/489 [00:31<00:38, 2.50it/s] Loading 0: 82%|████████▏ | 400/489 [00:32<00:21, 4.21it/s] Loading 0: 83%|████████▎ | 404/489 [00:32<00:15, 5.43it/s] Loading 0: 84%|████████▍ | 410/489 [00:32<00:09, 7.97it/s] Loading 0: 85%|████████▍ | 415/489 [00:32<00:07, 10.45it/s] Loading 0: 86%|████████▌ | 420/489 [00:32<00:05, 13.45it/s] Loading 0: 87%|████████▋ | 425/489 [00:32<00:03, 17.06it/s] Loading 0: 88%|████████▊ | 430/489 [00:32<00:03, 18.51it/s] Loading 0: 89%|████████▉ | 437/489 [00:32<00:02, 25.29it/s] Loading 0: 90%|█████████ | 442/489 [00:33<00:01, 28.27it/s] Loading 0: 91%|█████████▏| 447/489 [00:33<00:02, 20.84it/s] Loading 0: 93%|█████████▎| 454/489 [00:33<00:01, 27.47it/s] Loading 0: 94%|█████████▍| 459/489 [00:33<00:00, 30.03it/s] Loading 0: 95%|█████████▍| 464/489 [00:33<00:00, 32.24it/s] Loading 0: 96%|█████████▌| 469/489 [00:33<00:00, 34.90it/s] Loading 0: 97%|█████████▋| 474/489 [00:34<00:00, 30.40it/s] Loading 0: 98%|█████████▊| 481/489 [00:34<00:00, 36.82it/s] Loading 0: 99%|█████████▉| 486/489 [00:34<00:00, 37.45it/s]
Job yamatazen-ayla-light-16b-v1-mkmlizer completed after 137.06s with status: succeeded
Stopping job with name yamatazen-ayla-light-16b-v1-mkmlizer
Pipeline stage MKMLizer completed in 139.15s
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 yamatazen-ayla-light-16b-v1
Waiting for inference service yamatazen-ayla-light-16b-v1 to be ready
Inference service yamatazen-ayla-light-16b-v1 ready after 281.6505858898163s
Pipeline stage MKMLDeployer completed in 282.23s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.364511013031006s
Received healthy response to inference request in 1.8275721073150635s
Received healthy response to inference request in 1.8491144180297852s
Received healthy response to inference request in 2.0484111309051514s
Received healthy response to inference request in 1.8550560474395752s
5 requests
0 failed requests
5th percentile: 1.8318805694580078
10th percentile: 1.8361890316009521
20th percentile: 1.8448059558868408
30th percentile: 1.8503027439117432
40th percentile: 1.852679395675659
50th percentile: 1.8550560474395752
60th percentile: 1.9323980808258057
70th percentile: 2.009740114212036
80th percentile: 2.1116311073303224
90th percentile: 2.238071060180664
95th percentile: 2.301291036605835
Failed to get response for submission nischaydnk-mistral24bas_99124_v4: HTTPConnectionPool(host='nischaydnk-mistral24bas-99124-v4-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
99th percentile: 2.3518670177459717
mean time: 1.9889329433441163
Pipeline stage StressChecker completed in 11.35s
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 0.62s
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 0.62s
Shutdown handler de-registered
yamatazen-ayla-light-16b_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
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 5298.74s
Shutdown handler de-registered
yamatazen-ayla-light-16b_v1 status is now inactive due to auto deactivation removed underperforming models
yamatazen-ayla-light-16b_v1 status is now torndown due to DeploymentManager action