rica40325-feedback-and-dpo

developer_uid: rica40325

submission_id: rica40325-feedback-and-dpo_v2

model_name: rica40325-feedback-and-dpo_v1

model_group: rica40325/feedback-and-d

status: torndown

timestamp: 2024-09-01T08:12:40+00:00

num_battles: 12651

num_wins: 6343

celo_rating: 1233.31

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/feedback-and-dpo

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: rica40325-feedback-and-dpo_v1

is_internal_developer: False

language_model: rica40325/feedback-and-dpo

model_size: 8B

ranking_group: single

us_pacific_date: 2024-09-01

win_ratio: 0.5013832898585092

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, '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 rica40325-feedback-and-dpo-v2-mkmlizer
Waiting for job on rica40325-feedback-and-dpo-v2-mkmlizer to finish
rica40325-feedback-and-dpo-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-and-dpo-v2-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-and-dpo-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-and-dpo-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-and-dpo-v2-mkmlizer: Downloaded to shared memory in 34.263s
rica40325-feedback-and-dpo-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmptw_q9ldo, device:0
rica40325-feedback-and-dpo-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-and-dpo-v2-mkmlizer: quantized model in 25.940s
rica40325-feedback-and-dpo-v2-mkmlizer: Processed model rica40325/feedback-and-dpo in 60.204s
rica40325-feedback-and-dpo-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-and-dpo-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-and-dpo-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2
rica40325-feedback-and-dpo-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2/config.json
rica40325-feedback-and-dpo-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2/special_tokens_map.json
rica40325-feedback-and-dpo-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2/tokenizer_config.json
rica40325-feedback-and-dpo-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2/tokenizer.json
rica40325-feedback-and-dpo-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-and-dpo-v2/flywheel_model.0.safetensors
rica40325-feedback-and-dpo-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:06, 45.71it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 62.79it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 69.51it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:03, 75.34it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:03, 72.97it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:03, 74.66it/s]
Loading 0:  21%|██        | 61/291 [00:00<00:03, 74.52it/s]
Loading 0:  24%|██▍       | 70/291 [00:00<00:02, 75.19it/s]
Loading 0:  27%|██▋       | 79/291 [00:01<00:02, 77.68it/s]
Loading 0:  30%|██▉       | 87/291 [00:02<00:09, 21.48it/s]
Loading 0:  33%|███▎      | 97/291 [00:02<00:06, 27.89it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:05, 35.10it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 48.31it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:02, 55.03it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 61.03it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 65.46it/s]
Loading 0:  54%|█████▍    | 157/291 [00:02<00:01, 68.58it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 67.04it/s]
Loading 0:  60%|██████    | 175/291 [00:03<00:01, 71.69it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 71.93it/s]
Loading 0:  66%|██████▌   | 192/291 [00:04<00:04, 21.79it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 28.04it/s]
Loading 0:  74%|███████▎  | 214/291 [00:04<00:02, 36.69it/s]
Loading 0:  77%|███████▋  | 223/291 [00:04<00:01, 43.53it/s]
Loading 0:  80%|███████▉  | 232/291 [00:04<00:01, 48.53it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 55.09it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 59.14it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 62.38it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 64.33it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 69.79it/s]
Loading 0:  98%|█████████▊| 286/291 [00:05<00:00, 72.35it/s]
                                                            
Job rica40325-feedback-and-dpo-v2-mkmlizer completed after 83.7s with status: succeeded
Stopping job with name rica40325-feedback-and-dpo-v2-mkmlizer
Pipeline stage MKMLizer completed in 84.98s
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 rica40325-feedback-and-dpo-v2
Waiting for inference service rica40325-feedback-and-dpo-v2 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Failed to get response for submission blend_filor_2024-08-16: ('http://neversleep-noromaid-v0-8068-v149-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:41422->127.0.0.1:8080: read: connection reset by peer\n')
Failed to get response for submission chaiml-elo-alignment-run-3_v5: ('http://chaiml-elo-alignment-run-3-v5-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47136->127.0.0.1:8080: read: connection reset by peer\n')
Inference service rica40325-feedback-and-dpo-v2 ready after 291.6219575405121s
Pipeline stage MKMLDeployer completed in 291.98s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.239985227584839s
Received healthy response to inference request in 2.1172549724578857s
Received healthy response to inference request in 2.291229248046875s
Received healthy response to inference request in 1.6869244575500488s
Received healthy response to inference request in 1.5903294086456299s
5 requests
0 failed requests
5th percentile: 1.6096484184265136
10th percentile: 1.6289674282073974
20th percentile: 1.667605447769165
30th percentile: 1.7729905605316163
40th percentile: 1.945122766494751
50th percentile: 2.1172549724578857
60th percentile: 2.166347074508667
70th percentile: 2.2154391765594483
80th percentile: 2.250234031677246
90th percentile: 2.2707316398620607
95th percentile: 2.280980443954468
99th percentile: 2.2891794872283935
mean time: 1.9851446628570557
Pipeline stage StressChecker completed in 10.69s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
starting trigger_guanaco_pipeline %s
Pipeline stage TriggerMKMLProfilingPipeline completed in 4.39s
rica40325-feedback-and-dpo_v2 status is now deployed due to DeploymentManager action
run pipeline %s
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 rica40325-feedback-and-dpo-v2-profiler
Waiting for inference service rica40325-feedback-and-dpo-v2-profiler to be ready
Inference service rica40325-feedback-and-dpo-v2-profiler ready after 190.55445981025696s
Pipeline stage MKMLProfilerDeployer completed in 190.94s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
script pods %s
Pipeline stage MKMLProfilerRunner completed in 0.39s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-and-dpo-v2-profiler is running
Tearing down inference service rica40325-feedback-and-dpo-v2-profiler
Service rica40325-feedback-and-dpo-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.89s
rica40325-feedback-and-dpo_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-and-dpo_v2 status is now torndown due to DeploymentManager action