developer_uid: zonemercy
submission_id: chaiml-horror-sft-v1-5e4_v1
model_name: horror-sft
model_group: ChaiML/Horror-SFT-v1-5e4
status: torndown
timestamp: 2024-09-04T18:42:08+00:00
num_battles: 12753
num_wins: 5896
celo_rating: 1212.71
family_friendly_score: 0.0
submission_type: basic
model_repo: ChaiML/Horror-SFT-v1-5e4
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 4
max_input_tokens: 1024
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6409826641969656, 'latency_mean': 1.5600443160533906, 'latency_p50': 1.5565329790115356, 'latency_p90': 1.7208970785140991}, {'batch_size': 3, 'throughput': 1.2608907595656351, 'latency_mean': 2.3740212416648863, 'latency_p50': 2.3893237113952637, 'latency_p90': 2.6053009986877442}, {'batch_size': 5, 'throughput': 1.560240684537715, 'latency_mean': 3.189819926023483, 'latency_p50': 3.2029170989990234, 'latency_p90': 3.5496181726455687}, {'batch_size': 6, 'throughput': 1.6481788187630813, 'latency_mean': 3.607264202833176, 'latency_p50': 3.6057424545288086, 'latency_p90': 4.091022276878357}, {'batch_size': 8, 'throughput': 1.784597090106233, 'latency_mean': 4.452886284589767, 'latency_p50': 4.430305480957031, 'latency_p90': 5.08816499710083}, {'batch_size': 10, 'throughput': 1.8201333351521776, 'latency_mean': 5.444430181980133, 'latency_p50': 5.440035462379456, 'latency_p90': 6.087368774414062}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: horror-sft
is_internal_developer: True
language_model: ChaiML/Horror-SFT-v1-5e4
model_size: 13B
ranking_group: single
throughput_3p7s: 1.68
us_pacific_date: 2024-09-04
win_ratio: 0.46232259076295773
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
Resubmit model
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-horror-sft-v1-5e4-v1-mkmlizer
Waiting for job on chaiml-horror-sft-v1-5e4-v1-mkmlizer to finish
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ _____ __ __ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ /___/ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ Version: 0.10.1 ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ https://mk1.ai ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ belonging to: ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ Chai Research Corp. ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ║ ║
chaiml-horror-sft-v1-5e4-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission blend_jerun_2024-08-22: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"ValueError : [TypeError(\\"\'numpy.int64\' object is not iterable\\"), TypeError(\'vars() argument must have __dict__ attribute\')]"}')
chaiml-horror-sft-v1-5e4-v1-mkmlizer: Downloaded to shared memory in 52.330s
chaiml-horror-sft-v1-5e4-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpfeh40tw3, device:0
chaiml-horror-sft-v1-5e4-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-horror-sft-v1-5e4-v1-mkmlizer: quantized model in 35.848s
chaiml-horror-sft-v1-5e4-v1-mkmlizer: Processed model ChaiML/Horror-SFT-v1-5e4 in 88.179s
chaiml-horror-sft-v1-5e4-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-horror-sft-v1-5e4-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-horror-sft-v1-5e4-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v1
chaiml-horror-sft-v1-5e4-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v1/config.json
chaiml-horror-sft-v1-5e4-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v1/special_tokens_map.json
chaiml-horror-sft-v1-5e4-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v1/tokenizer_config.json
chaiml-horror-sft-v1-5e4-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v1/tokenizer.json
chaiml-horror-sft-v1-5e4-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:12, 29.46it/s] Loading 0: 4%|▎ | 13/363 [00:00<00:06, 50.95it/s] Loading 0: 5%|▌ | 19/363 [00:00<00:07, 46.44it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:07, 44.61it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:06, 50.41it/s] Loading 0: 10%|█ | 37/363 [00:00<00:07, 43.37it/s] Loading 0: 12%|█▏ | 42/363 [00:00<00:07, 42.71it/s] Loading 0: 13%|█▎ | 49/363 [00:01<00:06, 47.32it/s] Loading 0: 15%|█▍ | 54/363 [00:01<00:06, 47.66it/s] Loading 0: 17%|█▋ | 60/363 [00:01<00:06, 45.98it/s] Loading 0: 18%|█▊ | 65/363 [00:01<00:09, 31.34it/s] Loading 0: 20%|█▉ | 72/363 [00:01<00:07, 38.16it/s] Loading 0: 21%|██▏ | 78/363 [00:01<00:07, 39.61it/s] Loading 0: 23%|██▎ | 83/363 [00:01<00:07, 38.60it/s] Loading 0: 25%|██▍ | 90/363 [00:02<00:06, 43.94it/s] Loading 0: 26%|██▌ | 95/363 [00:02<00:05, 45.35it/s] Loading 0: 28%|██▊ | 100/363 [00:02<00:07, 36.15it/s] Loading 0: 30%|███ | 109/363 [00:02<00:05, 47.80it/s] Loading 0: 32%|███▏ | 115/363 [00:02<00:05, 46.49it/s] Loading 0: 33%|███▎ | 121/363 [00:02<00:05, 46.86it/s] Loading 0: 35%|███▍ | 127/363 [00:02<00:05, 42.45it/s] Loading 0: 37%|███▋ | 134/363 [00:03<00:04, 47.93it/s] Loading 0: 39%|███▊ | 140/363 [00:03<00:04, 48.30it/s] Loading 0: 40%|████ | 146/363 [00:03<00:06, 33.25it/s] Loading 0: 42%|████▏ | 151/363 [00:03<00:06, 33.81it/s] Loading 0: 43%|████▎ | 157/363 [00:03<00:05, 38.32it/s] Loading 0: 45%|████▍ | 163/363 [00:03<00:05, 37.78it/s] Loading 0: 46%|████▋ | 168/363 [00:04<00:05, 38.16it/s] Loading 0: 48%|████▊ | 175/363 [00:04<00:04, 43.84it/s] Loading 0: 50%|████▉ | 180/363 [00:04<00:04, 45.09it/s] Loading 0: 51%|█████ | 185/363 [00:04<00:04, 36.13it/s] Loading 0: 53%|█████▎ | 192/363 [00:04<00:03, 43.02it/s] Loading 0: 54%|█████▍ | 197/363 [00:04<00:03, 43.06it/s] Loading 0: 56%|█████▌ | 202/363 [00:04<00:03, 43.43it/s] Loading 0: 57%|█████▋ | 207/363 [00:04<00:03, 45.08it/s] Loading 0: 58%|█████▊ | 212/363 [00:05<00:04, 35.73it/s] Loading 0: 60%|██████ | 218/363 [00:05<00:03, 39.89it/s] Loading 0: 61%|██████▏ | 223/363 [00:05<00:04, 32.40it/s] Loading 0: 63%|██████▎ | 227/363 [00:05<00:04, 32.74it/s] Loading 0: 64%|██████▎ | 231/363 [00:05<00:04, 31.32it/s] Loading 0: 65%|██████▌ | 237/363 [00:05<00:03, 36.91it/s] Loading 0: 66%|██████▋ | 241/363 [00:05<00:03, 36.76it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:02, 40.83it/s] Loading 0: 69%|██████▉ | 252/363 [00:06<00:02, 42.11it/s] Loading 0: 71%|███████ | 257/363 [00:06<00:02, 35.85it/s] Loading 0: 73%|███████▎ | 265/363 [00:06<00:02, 44.01it/s] Loading 0: 75%|███████▍ | 271/363 [00:06<00:02, 42.16it/s] Loading 0: 76%|███████▌ | 276/363 [00:06<00:02, 41.19it/s] Loading 0: 78%|███████▊ | 283/363 [00:06<00:01, 46.15it/s] Loading 0: 79%|███████▉ | 288/363 [00:07<00:01, 45.10it/s] Loading 0: 81%|████████ | 293/363 [00:07<00:01, 36.38it/s] Loading 0: 82%|████████▏ | 299/363 [00:07<00:01, 40.48it/s] Loading 0: 84%|████████▎ | 304/363 [00:14<00:22, 2.57it/s] Loading 0: 85%|████████▍ | 308/363 [00:14<00:16, 3.30it/s] Loading 0: 86%|████████▌ | 312/363 [00:14<00:11, 4.28it/s] Loading 0: 88%|████████▊ | 320/363 [00:14<00:06, 7.11it/s] Loading 0: 90%|████████▉ | 326/363 [00:14<00:03, 9.53it/s] Loading 0: 91%|█████████ | 331/363 [00:14<00:02, 12.04it/s] Loading 0: 93%|█████████▎| 338/363 [00:14<00:01, 16.92it/s] Loading 0: 95%|█████████▍| 344/363 [00:14<00:00, 19.99it/s] Loading 0: 96%|█████████▌| 349/363 [00:15<00:00, 23.10it/s] Loading 0: 98%|█████████▊| 356/363 [00:15<00:00, 29.44it/s] Loading 0: 99%|█████████▉| 361/363 [00:15<00:00, 32.82it/s]
Job chaiml-horror-sft-v1-5e4-v1-mkmlizer completed after 115.3s with status: succeeded
Stopping job with name chaiml-horror-sft-v1-5e4-v1-mkmlizer
Pipeline stage MKMLizer completed in 116.58s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-horror-sft-v1-5e4-v1
Waiting for inference service chaiml-horror-sft-v1-5e4-v1 to be ready
Inference service chaiml-horror-sft-v1-5e4-v1 ready after 151.43829584121704s
Pipeline stage MKMLDeployer completed in 151.81s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9846713542938232s
Received healthy response to inference request in 1.1175541877746582s
Received healthy response to inference request in 1.3815388679504395s
Received healthy response to inference request in 1.9511737823486328s
Received healthy response to inference request in 1.7230703830718994s
5 requests
0 failed requests
5th percentile: 1.1703511238098145
10th percentile: 1.2231480598449707
20th percentile: 1.3287419319152831
30th percentile: 1.4498451709747315
40th percentile: 1.5864577770233155
50th percentile: 1.7230703830718994
60th percentile: 1.8143117427825928
70th percentile: 1.9055531024932861
80th percentile: 1.9578732967376709
90th percentile: 1.9712723255157472
95th percentile: 1.9779718399047852
99th percentile: 1.9833314514160156
mean time: 1.6316017150878905
Pipeline stage StressChecker completed in 9.07s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
starting trigger_guanaco_pipeline %s
Pipeline stage TriggerMKMLProfilingPipeline completed in 6.37s
chaiml-horror-sft-v1-5e4_v1 status is now deployed due to DeploymentManager action
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 0.11s
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 chaiml-horror-sft-v1-5e4-v1-profiler
Waiting for inference service chaiml-horror-sft-v1-5e4-v1-profiler to be ready
Inference service chaiml-horror-sft-v1-5e4-v1-profiler ready after 150.3678755760193s
Pipeline stage MKMLProfilerDeployer completed in 150.76s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-horror-sft-v1-5e4-v1-profiler-predictor-00001-deplo2jwbd:/code/chaiverse_profiler_1725475807 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-horror-sft-v1-5e4-v1-profiler-predictor-00001-deplo2jwbd --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725475807 && python profiles.py profile --best_of_n 4 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1725475807/summary.json'
Received signal 15, running shutdown handler
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e4-v1-profiler is running
Tearing down inference service chaiml-horror-sft-v1-5e4-v1-profiler
Service chaiml-horror-sft-v1-5e4-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.68s
Shutdown handler unregistered and original handlers restored
chaiml-horror-sft-v1-5e4_v1 status is now inactive due to auto deactivation removed underperforming models
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e4-v1-profiler is running
Skipping teardown as no inference service was found
Pipeline stage MKMLProfilerDeleter completed in 1.58s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-horror-sft-v1-5e4-v1-profiler
Waiting for inference service chaiml-horror-sft-v1-5e4-v1-profiler to be ready
Inference service chaiml-horror-sft-v1-5e4-v1-profiler ready after 150.3591010570526s
Pipeline stage MKMLProfilerDeployer completed in 150.69s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-horror-sft-v1-5e4-v1-profiler-predictor-00001-deplohhxch:/code/chaiverse_profiler_1725485346 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-horror-sft-v1-5e4-v1-profiler-predictor-00001-deplohhxch --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725485346 && python profiles.py profile --best_of_n 4 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1725485346/summary.json'
kubectl exec -it chaiml-horror-sft-v1-5e4-v1-profiler-predictor-00001-deplohhxch --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725485346/summary.json'
Pipeline stage MKMLProfilerRunner completed in 948.63s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e4-v1-profiler is running
Tearing down inference service chaiml-horror-sft-v1-5e4-v1-profiler
Service chaiml-horror-sft-v1-5e4-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.67s
Shutdown handler de-registered
chaiml-horror-sft-v1-5e4_v1 status is now torndown due to DeploymentManager action