yamatazen-ayla-light-12b

developer_uid: azuruce

submission_id: yamatazen-ayla-light-12b_v1

model_name: yamatazen-ayla-light-12b_v1

model_group: yamatazen/Ayla-Light-12B

status: torndown

timestamp: 2025-02-27T01:01:16+00:00

num_battles: 5978

num_wins: 2966

celo_rating: 1252.89

family_friendly_score: 0.5626

family_friendly_standard_error: 0.0070154292812343285

submission_type: basic

model_repo: yamatazen/Ayla-Light-12B

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6094266647907643, 'latency_mean': 1.6408148753643035, 'latency_p50': 1.633197546005249, 'latency_p90': 1.8053181648254395}, {'batch_size': 3, 'throughput': 1.1142976190009373, 'latency_mean': 2.6862270629405973, 'latency_p50': 2.695313811302185, 'latency_p90': 3.0010786771774294}, {'batch_size': 5, 'throughput': 1.337575169648541, 'latency_mean': 3.7167933213710787, 'latency_p50': 3.743143320083618, 'latency_p90': 4.186997747421264}, {'batch_size': 6, 'throughput': 1.4163077834085251, 'latency_mean': 4.223406983613968, 'latency_p50': 4.2356812953948975, 'latency_p90': 4.71100697517395}, {'batch_size': 8, 'throughput': 1.4708430480460806, 'latency_mean': 5.399167453050613, 'latency_p50': 5.415273666381836, 'latency_p90': 6.022688698768616}, {'batch_size': 10, 'throughput': 1.5225368987062644, 'latency_mean': 6.510112369060517, 'latency_p50': 6.505411148071289, 'latency_p90': 7.335489010810852}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: yamatazen-ayla-light-12b_v1

is_internal_developer: False

language_model: yamatazen/Ayla-Light-12B

model_size: 13B

ranking_group: single

throughput_3p7s: 1.34

us_pacific_date: 2025-02-26

win_ratio: 0.4961525593844095

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': ['###', 'User:', '####', 'Me:', '</s>', '<|im_end|>', '\n', '<|eot_id|>', '<|end_of_text|>', 'Bot:', 'You:'], '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-12b-v1-mkmlizer
Waiting for job on yamatazen-ayla-light-12b-v1-mkmlizer to finish
yamatazen-ayla-light-12b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
yamatazen-ayla-light-12b-v1-mkmlizer: ║     _____            __           __                                ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║       /___/                                                         ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║                                                                     ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  https://mk1.ai                                                     ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║                                                                     ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  belonging to:                                                      ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║                                                                     ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  Chai Research Corp.                                                ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
yamatazen-ayla-light-12b-v1-mkmlizer: ║                                                                     ║
yamatazen-ayla-light-12b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
yamatazen-ayla-light-12b-v1-mkmlizer: Downloaded to shared memory in 52.310s
yamatazen-ayla-light-12b-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpgzhxette, device:0
yamatazen-ayla-light-12b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
yamatazen-ayla-light-12b-v1-mkmlizer: quantized model in 38.918s
yamatazen-ayla-light-12b-v1-mkmlizer: Processed model yamatazen/Ayla-Light-12B in 91.230s
yamatazen-ayla-light-12b-v1-mkmlizer: creating bucket guanaco-mkml-models
yamatazen-ayla-light-12b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
yamatazen-ayla-light-12b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/yamatazen-ayla-light-12b-v1
yamatazen-ayla-light-12b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/yamatazen-ayla-light-12b-v1/config.json
yamatazen-ayla-light-12b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/yamatazen-ayla-light-12b-v1/special_tokens_map.json
yamatazen-ayla-light-12b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/yamatazen-ayla-light-12b-v1/tokenizer_config.json
yamatazen-ayla-light-12b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/yamatazen-ayla-light-12b-v1/flywheel_model.0.safetensors
yamatazen-ayla-light-12b-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/363 [00:06<18:48,  3.12s/it]
Loading 0:   2%|▏         | 6/363 [00:06<05:01,  1.18it/s]
Loading 0:   3%|▎         | 11/363 [00:06<02:12,  2.65it/s]
Loading 0:   4%|▍         | 15/363 [00:06<01:24,  4.13it/s]
Loading 0:   6%|▌         | 22/363 [00:06<00:44,  7.67it/s]
Loading 0:   7%|▋         | 26/363 [00:06<00:34,  9.74it/s]
Loading 0:   9%|▊         | 31/363 [00:07<00:25, 13.25it/s]
Loading 0:  10%|▉         | 35/363 [00:07<00:21, 15.48it/s]
Loading 0:  11%|█         | 39/363 [00:07<00:17, 18.50it/s]
Loading 0:  12%|█▏        | 43/363 [00:07<00:20, 15.30it/s]
Loading 0:  13%|█▎        | 47/363 [00:07<00:16, 18.67it/s]
Loading 0:  14%|█▍        | 51/363 [00:07<00:15, 20.47it/s]
Loading 0:  16%|█▌        | 58/363 [00:08<00:10, 28.15it/s]
Loading 0:  17%|█▋        | 62/363 [00:08<00:10, 29.19it/s]
Loading 0:  18%|█▊        | 67/363 [00:08<00:09, 32.74it/s]
Loading 0:  20%|█▉        | 71/363 [00:08<00:09, 32.24it/s]
Loading 0:  21%|██        | 76/363 [00:08<00:08, 35.65it/s]
Loading 0:  22%|██▏       | 80/363 [00:08<00:08, 34.67it/s]
Loading 0:  23%|██▎       | 85/363 [00:08<00:07, 36.74it/s]
Loading 0:  25%|██▍       | 89/363 [00:08<00:07, 35.05it/s]
Loading 0:  26%|██▌       | 94/363 [00:09<00:07, 37.27it/s]
Loading 0:  27%|██▋       | 98/363 [00:09<00:07, 35.38it/s]
Loading 0:  28%|██▊       | 103/363 [00:09<00:07, 37.14it/s]
Loading 0:  29%|██▉       | 107/363 [00:09<00:07, 34.02it/s]
Loading 0:  31%|███       | 112/363 [00:09<00:06, 36.37it/s]
Loading 0:  32%|███▏      | 116/363 [00:09<00:07, 35.13it/s]
Loading 0:  33%|███▎      | 121/363 [00:09<00:09, 25.57it/s]
Loading 0:  34%|███▍      | 124/363 [00:10<00:09, 25.51it/s]
Loading 0:  36%|███▌      | 130/363 [00:10<00:07, 31.15it/s]
Loading 0:  37%|███▋      | 134/363 [00:10<00:07, 31.34it/s]
Loading 0:  38%|███▊      | 139/363 [00:10<00:06, 33.98it/s]
Loading 0:  39%|███▉      | 143/363 [00:10<00:06, 33.67it/s]
Loading 0:  41%|████      | 148/363 [00:10<00:05, 35.94it/s]
Loading 0:  42%|████▏     | 152/363 [00:10<00:06, 34.19it/s]
Loading 0:  43%|████▎     | 156/363 [00:10<00:05, 35.61it/s]
Loading 0:  44%|████▍     | 160/363 [00:11<00:06, 32.85it/s]
Loading 0:  46%|████▌     | 166/363 [00:11<00:05, 38.20it/s]
Loading 0:  47%|████▋     | 170/363 [00:11<00:05, 37.20it/s]
Loading 0:  48%|████▊     | 175/363 [00:11<00:04, 38.94it/s]
Loading 0:  49%|████▉     | 179/363 [00:11<00:04, 37.92it/s]
Loading 0:  51%|█████     | 184/363 [00:11<00:04, 40.79it/s]
Loading 0:  52%|█████▏    | 189/363 [00:11<00:04, 40.75it/s]
Loading 0:  53%|█████▎    | 194/363 [00:11<00:04, 39.58it/s]
Loading 0:  55%|█████▍    | 198/363 [00:12<00:04, 35.58it/s]
Loading 0:  56%|█████▌    | 202/363 [00:12<00:06, 23.63it/s]
Loading 0:  56%|█████▋    | 205/363 [00:12<00:06, 23.42it/s]
Loading 0:  58%|█████▊    | 211/363 [00:12<00:05, 29.74it/s]
Loading 0:  59%|█████▉    | 215/363 [00:12<00:04, 30.98it/s]
Loading 0:  61%|██████    | 220/363 [00:12<00:04, 33.64it/s]
Loading 0:  62%|██████▏   | 224/363 [00:12<00:04, 32.52it/s]
Loading 0:  63%|██████▎   | 229/363 [00:13<00:03, 34.69it/s]
Loading 0:  64%|██████▍   | 233/363 [00:13<00:03, 33.50it/s]
Loading 0:  66%|██████▌   | 238/363 [00:13<00:03, 35.61it/s]
Loading 0:  67%|██████▋   | 242/363 [00:13<00:03, 34.41it/s]
Loading 0:  68%|██████▊   | 247/363 [00:13<00:03, 37.06it/s]
Loading 0:  69%|██████▉   | 251/363 [00:13<00:03, 36.04it/s]
Loading 0:  71%|███████   | 256/363 [00:13<00:02, 38.00it/s]
Loading 0:  72%|███████▏  | 260/363 [00:13<00:03, 33.54it/s]
Loading 0:  73%|███████▎  | 264/363 [00:14<00:02, 33.65it/s]
Loading 0:  74%|███████▍  | 268/363 [00:14<00:03, 31.28it/s]
Loading 0:  75%|███████▌  | 274/363 [00:14<00:02, 35.56it/s]
Loading 0:  77%|███████▋  | 278/363 [00:14<00:02, 34.15it/s]
Loading 0:  78%|███████▊  | 283/363 [00:14<00:03, 24.66it/s]
Loading 0:  79%|███████▉  | 286/363 [00:14<00:03, 24.80it/s]
Loading 0:  80%|████████  | 292/363 [00:15<00:02, 30.69it/s]
Loading 0:  82%|████████▏ | 296/363 [00:15<00:02, 31.36it/s]
Loading 0:  83%|████████▎ | 301/363 [00:15<00:01, 34.31it/s]
Loading 0:  84%|████████▍ | 305/363 [00:15<00:01, 33.12it/s]
Loading 0:  85%|████████▌ | 310/363 [00:15<00:01, 35.10it/s]
Loading 0:  87%|████████▋ | 314/363 [00:15<00:01, 33.76it/s]
Loading 0:  88%|████████▊ | 319/363 [00:15<00:01, 35.99it/s]
Loading 0:  89%|████████▉ | 323/363 [00:15<00:01, 34.60it/s]
Loading 0:  90%|█████████ | 328/363 [00:16<00:00, 36.28it/s]
Loading 0:  91%|█████████▏| 332/363 [00:16<00:00, 34.68it/s]
Loading 0:  93%|█████████▎| 337/363 [00:16<00:00, 36.58it/s]
Loading 0:  94%|█████████▍| 341/363 [00:16<00:00, 34.75it/s]
Loading 0:  95%|█████████▌| 346/363 [00:16<00:00, 35.99it/s]
Loading 0:  96%|█████████▋| 350/363 [00:16<00:00, 34.05it/s]
Loading 0:  98%|█████████▊| 355/363 [00:16<00:00, 35.22it/s]
Loading 0:  99%|█████████▉| 359/363 [00:16<00:00, 33.31it/s]
                                                            
Job yamatazen-ayla-light-12b-v1-mkmlizer completed after 130.39s with status: succeeded
Stopping job with name yamatazen-ayla-light-12b-v1-mkmlizer
Pipeline stage MKMLizer completed in 130.96s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.17s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service yamatazen-ayla-light-12b-v1
Waiting for inference service yamatazen-ayla-light-12b-v1 to be ready
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)
Unable to record family friendly update due to error: ('http://chaiml-nemo-guard-merged-v2-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:37926->127.0.0.1:8080: read: connection reset by peer\n')
Inference service yamatazen-ayla-light-12b-v1 ready after 292.54549169540405s
Pipeline stage MKMLDeployer completed in 294.15s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.5994462966918945s
Received healthy response to inference request in 1.4376115798950195s
Received healthy response to inference request in 1.456955909729004s
Received healthy response to inference request in 0.67207932472229s
Received healthy response to inference request in 1.0988879203796387s
5 requests
0 failed requests
5th percentile: 0.7574410438537598
10th percentile: 0.8428027629852295
20th percentile: 1.013526201248169
30th percentile: 1.1666326522827148
40th percentile: 1.3021221160888672
50th percentile: 1.4376115798950195
60th percentile: 1.4453493118286134
70th percentile: 1.453087043762207
80th percentile: 1.4854539871215822
90th percentile: 1.5424501419067382
95th percentile: 1.5709482192993163
99th percentile: 1.593746681213379
mean time: 1.2529962062835693
Pipeline stage StressChecker completed in 7.64s
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 0.69s
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 0.70s
Shutdown handler de-registered
yamatazen-ayla-light-12b_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 yamatazen-ayla-light-12b-v1-profiler
Waiting for inference service yamatazen-ayla-light-12b-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 2594.53s
Shutdown handler de-registered
yamatazen-ayla-light-12b_v1 status is now inactive due to auto deactivation removed underperforming models
yamatazen-ayla-light-12b_v1 status is now torndown due to DeploymentManager action