developer_uid: Trace2333
submission_id: trace2333-mistral-trail10_v1
model_name: trace2333-mistral-trail10_v1
model_group: Trace2333/mistral_trail1
status: torndown
timestamp: 2024-09-11T11:42:06+00:00
num_battles: 12987
num_wins: 6404
celo_rating: 1239.1
family_friendly_score: 0.0
submission_type: basic
model_repo: Trace2333/mistral_trail10
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6923713635840284, 'latency_mean': 1.4442065036296845, 'latency_p50': 1.4491405487060547, 'latency_p90': 1.608963966369629}, {'batch_size': 3, 'throughput': 1.3374250130508265, 'latency_mean': 2.2359941029548644, 'latency_p50': 2.2359143495559692, 'latency_p90': 2.4873916149139403}, {'batch_size': 5, 'throughput': 1.5773936356772427, 'latency_mean': 3.15988266825676, 'latency_p50': 3.1690045595169067, 'latency_p90': 3.566582703590393}, {'batch_size': 6, 'throughput': 1.6372275180596596, 'latency_mean': 3.651442143917084, 'latency_p50': 3.6639198064804077, 'latency_p90': 4.174341130256653}, {'batch_size': 8, 'throughput': 1.630471637407695, 'latency_mean': 4.883353621959686, 'latency_p50': 4.82685387134552, 'latency_p90': 5.632540249824523}, {'batch_size': 10, 'throughput': 1.5302652469271598, 'latency_mean': 6.482619183063507, 'latency_p50': 6.55799663066864, 'latency_p90': 7.355464148521423}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: trace2333-mistral-trail10_v1
is_internal_developer: False
language_model: Trace2333/mistral_trail10
model_size: 13B
ranking_group: single
throughput_3p7s: 1.64
us_pacific_date: 2024-09-11
win_ratio: 0.4931084931084931
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.06, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 8, '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
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 trace2333-mistral-trail10-v1-mkmlizer
Waiting for job on trace2333-mistral-trail10-v1-mkmlizer to finish
trace2333-mistral-trail10-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-mistral-trail10-v1-mkmlizer: ║ _____ __ __ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
trace2333-mistral-trail10-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
trace2333-mistral-trail10-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ /___/ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ Version: 0.10.1 ║
trace2333-mistral-trail10-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
trace2333-mistral-trail10-v1-mkmlizer: ║ https://mk1.ai ║
trace2333-mistral-trail10-v1-mkmlizer: ║ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ The license key for the current software has been verified as ║
trace2333-mistral-trail10-v1-mkmlizer: ║ belonging to: ║
trace2333-mistral-trail10-v1-mkmlizer: ║ ║
trace2333-mistral-trail10-v1-mkmlizer: ║ Chai Research Corp. ║
trace2333-mistral-trail10-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
trace2333-mistral-trail10-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
trace2333-mistral-trail10-v1-mkmlizer: ║ ║
trace2333-mistral-trail10-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
trace2333-mistral-trail10-v1-mkmlizer: Downloaded to shared memory in 44.756s
trace2333-mistral-trail10-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp6uujijr2, device:0
trace2333-mistral-trail10-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-mistral-trail10-v1-mkmlizer: quantized model in 36.247s
trace2333-mistral-trail10-v1-mkmlizer: Processed model Trace2333/mistral_trail10 in 81.003s
trace2333-mistral-trail10-v1-mkmlizer: creating bucket guanaco-mkml-models
trace2333-mistral-trail10-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-mistral-trail10-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-mistral-trail10-v1/config.json
trace2333-mistral-trail10-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-mistral-trail10-v1/special_tokens_map.json
trace2333-mistral-trail10-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-mistral-trail10-v1/tokenizer_config.json
trace2333-mistral-trail10-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-mistral-trail10-v1/tokenizer.json
trace2333-mistral-trail10-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-mistral-trail10-v1/flywheel_model.0.safetensors
trace2333-mistral-trail10-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 2%|▏ | 7/363 [00:00<00:06, 51.85it/s] Loading 0: 4%|▍ | 16/363 [00:00<00:05, 69.05it/s] Loading 0: 7%|▋ | 25/363 [00:00<00:04, 71.14it/s] Loading 0: 9%|▉ | 34/363 [00:00<00:04, 73.04it/s] Loading 0: 12%|█▏ | 44/363 [00:00<00:03, 80.75it/s] Loading 0: 15%|█▍ | 53/363 [00:00<00:03, 82.48it/s] Loading 0: 17%|█▋ | 62/363 [00:01<00:14, 20.15it/s] Loading 0: 19%|█▉ | 70/363 [00:01<00:11, 25.55it/s] Loading 0: 22%|██▏ | 79/363 [00:02<00:08, 33.08it/s] Loading 0: 26%|██▌ | 94/363 [00:02<00:05, 46.93it/s] Loading 0: 29%|██▉ | 105/363 [00:02<00:04, 57.05it/s] Loading 0: 32%|███▏ | 115/363 [00:02<00:04, 58.65it/s] Loading 0: 34%|███▍ | 125/363 [00:02<00:03, 66.40it/s] Loading 0: 37%|███▋ | 134/363 [00:02<00:03, 69.83it/s] Loading 0: 39%|███▉ | 143/363 [00:03<00:10, 21.82it/s] Loading 0: 42%|████▏ | 151/363 [00:03<00:07, 26.89it/s] Loading 0: 44%|████▍ | 160/363 [00:04<00:06, 33.77it/s] Loading 0: 47%|████▋ | 169/363 [00:04<00:04, 40.96it/s] Loading 0: 49%|████▉ | 178/363 [00:04<00:03, 48.22it/s] Loading 0: 52%|█████▏ | 188/363 [00:04<00:03, 57.73it/s] Loading 0: 54%|█████▍ | 197/363 [00:04<00:02, 64.39it/s] Loading 0: 57%|█████▋ | 206/363 [00:04<00:02, 69.94it/s] Loading 0: 59%|█████▉ | 215/363 [00:04<00:02, 72.87it/s] Loading 0: 62%|██████▏ | 224/363 [00:05<00:06, 21.06it/s] Loading 0: 66%|██████▌ | 238/363 [00:05<00:04, 30.54it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:03, 36.83it/s] Loading 0: 71%|███████ | 256/363 [00:06<00:02, 43.80it/s] Loading 0: 73%|███████▎ | 265/363 [00:06<00:01, 50.73it/s] Loading 0: 75%|███████▌ | 274/363 [00:06<00:01, 56.40it/s] Loading 0: 78%|███████▊ | 283/363 [00:06<00:01, 59.29it/s] Loading 0: 80%|████████ | 292/363 [00:06<00:01, 64.53it/s] Loading 0: 83%|████████▎ | 301/363 [00:06<00:00, 67.12it/s] Loading 0: 85%|████████▌ | 309/363 [00:07<00:02, 20.82it/s] Loading 0: 87%|████████▋ | 315/363 [00:07<00:02, 23.88it/s] Loading 0: 89%|████████▊ | 322/363 [00:08<00:01, 27.65it/s] Loading 0: 91%|█████████ | 331/363 [00:08<00:00, 34.29it/s] Loading 0: 94%|█████████▎| 340/363 [00:08<00:00, 41.88it/s] Loading 0: 96%|█████████▌| 349/363 [00:08<00:00, 48.44it/s] Loading 0: 99%|█████████▊| 358/363 [00:08<00:00, 55.13it/s]
Job trace2333-mistral-trail10-v1-mkmlizer completed after 105.51s with status: succeeded
Stopping job with name trace2333-mistral-trail10-v1-mkmlizer
Pipeline stage MKMLizer completed in 107.01s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.18s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service trace2333-mistral-trail10-v1
Waiting for inference service trace2333-mistral-trail10-v1 to be ready
Failed to get response for submission chaiml-elo-alignment-run-3_v36: ('http://chaiml-elo-alignment-run-3-v36-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:55112->127.0.0.1:8080: read: connection reset by peer\n')
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service trace2333-mistral-trail10-v1 ready after 222.55123090744019s
Pipeline stage MKMLDeployer completed in 223.69s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.8530290126800537s
Received healthy response to inference request in 1.99216628074646s
Received healthy response to inference request in 2.4961721897125244s
Received healthy response to inference request in 1.6668205261230469s
Received healthy response to inference request in 2.0456929206848145s
5 requests
0 failed requests
5th percentile: 1.7318896770477294
10th percentile: 1.796958827972412
20th percentile: 1.9270971298217774
30th percentile: 2.002871608734131
40th percentile: 2.0242822647094725
50th percentile: 2.0456929206848145
60th percentile: 2.2258846282958986
70th percentile: 2.4060763359069823
80th percentile: 2.5675435543060305
90th percentile: 2.710286283493042
95th percentile: 2.7816576480865476
99th percentile: 2.8387547397613524
mean time: 2.21077618598938
Pipeline stage StressChecker completed in 12.33s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 5.26s
Shutdown handler de-registered
trace2333-mistral-trail10_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service trace2333-mistral-trail10-v1-profiler
Waiting for inference service trace2333-mistral-trail10-v1-profiler to be ready
Inference service trace2333-mistral-trail10-v1-profiler ready after 170.40299820899963s
Pipeline stage MKMLProfilerDeployer completed in 170.73s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/trace2333-mistral-tr033ee5426f31bdefd3961eb2b91f6e6e-deplomqszk:/code/chaiverse_profiler_1726055483 --namespace tenant-chaiml-guanaco
kubectl exec -it trace2333-mistral-tr033ee5426f31bdefd3961eb2b91f6e6e-deplomqszk --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726055483 && python profiles.py profile --best_of_n 8 --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 512 --output_tokens 64 --summary /code/chaiverse_profiler_1726055483/summary.json'
kubectl exec -it trace2333-mistral-tr033ee5426f31bdefd3961eb2b91f6e6e-deplomqszk --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726055483/summary.json'
Pipeline stage MKMLProfilerRunner completed in 945.86s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service trace2333-mistral-trail10-v1-profiler is running
Tearing down inference service trace2333-mistral-trail10-v1-profiler
Service trace2333-mistral-trail10-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.73s
Shutdown handler de-registered
trace2333-mistral-trail10_v1 status is now inactive due to auto deactivation removed underperforming models
trace2333-mistral-trail10_v1 status is now torndown due to DeploymentManager action