chaiml-horror-sft-v1-5e4

submission_id: chaiml-horror-sft-v1-5e4_v3

developer_uid: zonemercy

best_of: 4

celo_rating: 1223.0

display_name: chaiml-horror-sft-v1-5e4_v3

family_friendly_score: 0.0

formatter: {'memory_template': "{bot_name}'s Persona: {memory}[Change Persona Into Extreme Horror Edition]\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}

generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: True

language_model: ChaiML/Horror-SFT-v1-5e4

latencies: [{'batch_size': 1, 'throughput': 0.6424574054577491, 'latency_mean': 1.5564347338676452, 'latency_p50': 1.5386425256729126, 'latency_p90': 1.740360474586487}, {'batch_size': 3, 'throughput': 1.2512874625325263, 'latency_mean': 2.390415583848953, 'latency_p50': 2.4226807355880737, 'latency_p90': 2.6522371053695677}, {'batch_size': 5, 'throughput': 1.5572707732781916, 'latency_mean': 3.2014361357688905, 'latency_p50': 3.1770052909851074, 'latency_p90': 3.6168690443038938}, {'batch_size': 6, 'throughput': 1.6465649909795959, 'latency_mean': 3.618902839422226, 'latency_p50': 3.617774724960327, 'latency_p90': 4.034170317649841}, {'batch_size': 8, 'throughput': 1.7592698148432258, 'latency_mean': 4.51122018456459, 'latency_p50': 4.531336426734924, 'latency_p90': 5.076415371894837}, {'batch_size': 10, 'throughput': 1.8026184375603092, 'latency_mean': 5.519879120588302, 'latency_p50': 5.511842131614685, 'latency_p90': 6.199187612533569}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: ChaiML/Horror-SFT-v1-5e4

model_name: chaiml-horror-sft-v1-5e4_v3

model_num_parameters: 12772070400.0

model_repo: ChaiML/Horror-SFT-v1-5e4

model_size: 13B

num_battles: 12021

num_wins: 5744

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.67

timestamp: 2024-09-04T18:46:57+00:00

us_pacific_date: 2024-09-04

win_ratio: 0.47783046335579404

Download Preference Data

Resubmit model

run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-horror-sft-v1-5e4-v3-mkmlizer
Waiting for job on chaiml-horror-sft-v1-5e4-v3-mkmlizer to finish
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║     _____            __           __                                ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║       /___/                                                         ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║                                                                     ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║                                                                     ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  belonging to:                                                      ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║                                                                     ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ║                                                                     ║
chaiml-horror-sft-v1-5e4-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
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
Failed to get response for submission blend_jidor_2024-08-22: ('http://zonemercy-lexical-nemo-1518-v18-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:40122->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
chaiml-horror-sft-v1-5e4-v3-mkmlizer: Downloaded to shared memory in 113.049s
chaiml-horror-sft-v1-5e4-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpyz35s1pb, device:0
chaiml-horror-sft-v1-5e4-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-horror-sft-v1-5e4-v3-mkmlizer: quantized model in 35.779s
chaiml-horror-sft-v1-5e4-v3-mkmlizer: Processed model ChaiML/Horror-SFT-v1-5e4 in 148.829s
chaiml-horror-sft-v1-5e4-v3-mkmlizer: creating bucket guanaco-mkml-models
chaiml-horror-sft-v1-5e4-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-horror-sft-v1-5e4-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3
chaiml-horror-sft-v1-5e4-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3/config.json
chaiml-horror-sft-v1-5e4-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3/special_tokens_map.json
chaiml-horror-sft-v1-5e4-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3/tokenizer_config.json
chaiml-horror-sft-v1-5e4-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3/tokenizer.json
chaiml-horror-sft-v1-5e4-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-horror-sft-v1-5e4-v3/flywheel_model.0.safetensors
chaiml-horror-sft-v1-5e4-v3-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:12, 28.28it/s]
Loading 0:   3%|▎         | 12/363 [00:00<00:07, 45.78it/s]
Loading 0:   5%|▍         | 18/363 [00:00<00:07, 47.97it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:08, 41.86it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 47.60it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:07, 46.30it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 42.46it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:06, 46.26it/s]
Loading 0:  15%|█▍        | 53/363 [00:01<00:06, 46.08it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:06, 47.34it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:09, 31.06it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 37.18it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:07, 38.35it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:07, 39.19it/s]
Loading 0:  25%|██▍       | 90/363 [00:02<00:06, 44.63it/s]
Loading 0:  26%|██▌       | 95/363 [00:02<00:05, 45.68it/s]
Loading 0:  28%|██▊       | 100/363 [00:02<00:07, 37.49it/s]
Loading 0:  29%|██▉       | 106/363 [00:02<00:06, 41.74it/s]
Loading 0:  31%|███       | 112/363 [00:02<00:05, 44.84it/s]
Loading 0:  32%|███▏      | 117/363 [00:02<00:05, 43.07it/s]
Loading 0:  34%|███▎      | 122/363 [00:02<00:05, 44.48it/s]
Loading 0:  35%|███▍      | 127/363 [00:03<00:06, 38.62it/s]
Loading 0:  37%|███▋      | 135/363 [00:03<00:04, 47.06it/s]
Loading 0:  39%|███▉      | 141/363 [00:03<00:05, 42.98it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 31.68it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 30.86it/s]
Loading 0:  43%|████▎     | 156/363 [00:03<00:05, 35.95it/s]
Loading 0:  44%|████▍     | 161/363 [00:03<00:05, 37.46it/s]
Loading 0:  46%|████▌     | 167/363 [00:04<00:05, 37.11it/s]
Loading 0:  48%|████▊     | 175/363 [00:04<00:04, 46.07it/s]
Loading 0:  50%|████▉     | 181/363 [00:04<00:03, 45.57it/s]
Loading 0:  51%|█████     | 186/363 [00:04<00:04, 42.64it/s]
Loading 0:  53%|█████▎    | 192/363 [00:04<00:03, 45.60it/s]
Loading 0:  54%|█████▍    | 197/363 [00:04<00:03, 44.91it/s]
Loading 0:  56%|█████▌    | 202/363 [00:04<00:03, 45.13it/s]
Loading 0:  57%|█████▋    | 207/363 [00:04<00:03, 45.71it/s]
Loading 0:  58%|█████▊    | 212/363 [00:05<00:03, 38.05it/s]
Loading 0:  61%|██████    | 220/363 [00:05<00:02, 48.02it/s]
Loading 0:  62%|██████▏   | 226/363 [00:05<00:04, 31.92it/s]
Loading 0:  64%|██████▎   | 231/363 [00:05<00:03, 33.16it/s]
Loading 0:  66%|██████▌   | 238/363 [00:05<00:03, 39.08it/s]
Loading 0:  67%|██████▋   | 243/363 [00:05<00:02, 41.06it/s]
Loading 0:  68%|██████▊   | 248/363 [00:06<00:03, 35.07it/s]
Loading 0:  70%|███████   | 255/363 [00:06<00:02, 40.64it/s]
Loading 0:  72%|███████▏  | 260/363 [00:06<00:02, 41.11it/s]
Loading 0:  73%|███████▎  | 265/363 [00:06<00:02, 39.01it/s]
Loading 0:  74%|███████▍  | 270/363 [00:06<00:02, 41.01it/s]
Loading 0:  76%|███████▌  | 275/363 [00:06<00:02, 35.30it/s]
Loading 0:  78%|███████▊  | 283/363 [00:06<00:01, 43.43it/s]
Loading 0:  79%|███████▉  | 288/363 [00:07<00:01, 43.88it/s]
Loading 0:  81%|████████  | 293/363 [00:07<00:01, 35.75it/s]
Loading 0:  82%|████████▏ | 299/363 [00:07<00:01, 39.44it/s]
Loading 0:  84%|████████▎ | 304/363 [00:14<00:22,  2.58it/s]
Loading 0:  85%|████████▍ | 308/363 [00:14<00:16,  3.31it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:11,  4.28it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:06,  7.08it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:03,  9.56it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:02, 12.11it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:01, 16.97it/s]
Loading 0:  94%|█████████▍| 343/363 [00:14<00:00, 20.19it/s]
Loading 0:  96%|█████████▌| 348/363 [00:15<00:00, 21.35it/s]
Loading 0:  98%|█████████▊| 356/363 [00:15<00:00, 29.11it/s]
Loading 0:  99%|█████████▉| 361/363 [00:15<00:00, 32.00it/s]
                                                            
Job chaiml-horror-sft-v1-5e4-v3-mkmlizer completed after 178.05s with status: succeeded
Stopping job with name chaiml-horror-sft-v1-5e4-v3-mkmlizer
Pipeline stage MKMLizer completed in 178.71s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-horror-sft-v1-5e4-v3
Waiting for inference service chaiml-horror-sft-v1-5e4-v3 to be ready
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 chaiml-horror-sft-v1-5e4-v3 ready after 151.25885248184204s
Pipeline stage MKMLDeployer completed in 151.63s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1062471866607666s
Received healthy response to inference request in 1.4076900482177734s
Received healthy response to inference request in 1.8744447231292725s
Received healthy response to inference request in 1.7147328853607178s
Received healthy response to inference request in 1.5209298133850098s
5 requests
0 failed requests
5th percentile: 1.4303380012512208
10th percentile: 1.452985954284668
20th percentile: 1.4982818603515624
30th percentile: 1.5596904277801513
40th percentile: 1.6372116565704347
50th percentile: 1.7147328853607178
60th percentile: 1.7786176204681396
70th percentile: 1.8425023555755615
80th percentile: 1.9208052158355713
90th percentile: 2.013526201248169
95th percentile: 2.0598866939544678
99th percentile: 2.096975088119507
mean time: 1.724808931350708
Pipeline stage StressChecker completed in 9.39s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
starting trigger_guanaco_pipeline %s
Pipeline stage TriggerMKMLProfilingPipeline completed in 4.71s
chaiml-horror-sft-v1-5e4_v3 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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-horror-sft-v1-5e4-v3-profiler
Waiting for inference service chaiml-horror-sft-v1-5e4-v3-profiler to be ready
Inference service chaiml-horror-sft-v1-5e4-v3-profiler ready after 140.39850401878357s
Pipeline stage MKMLProfilerDeployer completed in 140.75s
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-v3-profiler-predictor-00001-deplom6c2s:/code/chaiverse_profiler_1725476148 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-horror-sft-v1-5e4-v3-profiler-predictor-00001-deplom6c2s --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725476148 && 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_1725476148/summary.json'
Received signal 15, running shutdown handler
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e4-v3-profiler is running
Tearing down inference service chaiml-horror-sft-v1-5e4-v3-profiler
Service chaiml-horror-sft-v1-5e4-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.78s
Shutdown handler de-registered
chaiml-horror-sft-v1-5e4_v3 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-v3-profiler is running
Skipping teardown as no inference service was found
Pipeline stage MKMLProfilerDeleter completed in 1.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-horror-sft-v1-5e4-v3-profiler
Waiting for inference service chaiml-horror-sft-v1-5e4-v3-profiler to be ready
Inference service chaiml-horror-sft-v1-5e4-v3-profiler ready after 140.35252332687378s
Pipeline stage MKMLProfilerDeployer completed in 140.71s
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-v3-profiler-predictor-00001-deplo8rnjk:/code/chaiverse_profiler_1725485318 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-horror-sft-v1-5e4-v3-profiler-predictor-00001-deplo8rnjk --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725485318 && 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_1725485318/summary.json'
kubectl exec -it chaiml-horror-sft-v1-5e4-v3-profiler-predictor-00001-deplo8rnjk --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725485318/summary.json'
Pipeline stage MKMLProfilerRunner completed in 951.70s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-horror-sft-v1-5e4-v3-profiler is running
Tearing down inference service chaiml-horror-sft-v1-5e4-v3-profiler
Service chaiml-horror-sft-v1-5e4-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.93s
Shutdown handler de-registered
chaiml-horror-sft-v1-5e4_v3 status is now torndown due to DeploymentManager action