pawankrd-cosmosrp-llama31

submission_id: pawankrd-cosmosrp-llama31_v1

developer_uid: PawanOsman

best_of: 16

celo_rating: 1258.28

display_name: cosmosrp-llama31-test

family_friendly_score: 0.0

formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}

generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['<', '>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 128}

is_internal_developer: False

language_model: PawanKrd/cosmosrp-llama31

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: PawanKrd/cosmosrp-llama3

model_name: cosmosrp-llama31-test

model_num_parameters: 8030261248.0

model_repo: PawanKrd/cosmosrp-llama31

model_size: 8B

num_battles: 13643

num_wins: 7803

ranking_group: single

reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}

reward_repo: Jellywibble/gpt2_xl_pairwise_89m_step_347634

status: torndown

submission_type: basic

timestamp: 2024-07-26T12:57:08+00:00

us_pacific_date: 2024-07-26

win_ratio: 0.5719416550612035

Download Preference Data

Resubmit model

Running pipeline stage MKMLizer
Starting job with name pawankrd-cosmosrp-llama31-v1-mkmlizer
Waiting for job on pawankrd-cosmosrp-llama31-v1-mkmlizer to finish
pawankrd-cosmosrp-llama31-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║     _____            __           __                                ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║       /___/                                                         ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║                                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  Version: 0.9.7                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  https://mk1.ai                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║                                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  belonging to:                                                      ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║                                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  Chai Research Corp.                                                ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ║                                                                     ║
pawankrd-cosmosrp-llama31-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
pawankrd-cosmosrp-llama31-v1-mkmlizer: Downloaded to shared memory in 30.060s
pawankrd-cosmosrp-llama31-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmppzlu9sty, device:0
pawankrd-cosmosrp-llama31-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
pawankrd-cosmosrp-llama31-v1-mkmlizer: quantized model in 25.493s
pawankrd-cosmosrp-llama31-v1-mkmlizer: Processed model PawanKrd/cosmosrp-llama31 in 55.554s
pawankrd-cosmosrp-llama31-v1-mkmlizer: creating bucket guanaco-mkml-models
pawankrd-cosmosrp-llama31-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
pawankrd-cosmosrp-llama31-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/pawankrd-cosmosrp-llama31-v1
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/pawankrd-cosmosrp-llama31-v1/tokenizer.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/pawankrd-cosmosrp-llama31-v1/flywheel_model.0.safetensors
pawankrd-cosmosrp-llama31-v1-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
pawankrd-cosmosrp-llama31-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<11:33,  2.40s/it]
Loading 0:   2%|▏         | 6/291 [00:04<03:04,  1.55it/s]
Loading 0:   4%|▍         | 13/291 [00:05<01:05,  4.22it/s]
Loading 0:   7%|▋         | 20/291 [00:05<00:35,  7.54it/s]
Loading 0:   9%|▉         | 27/291 [00:05<00:22, 11.82it/s]
Loading 0:  11%|█▏        | 33/291 [00:05<00:16, 15.33it/s]
Loading 0:  14%|█▍        | 42/291 [00:05<00:11, 21.99it/s]
Loading 0:  18%|█▊        | 51/291 [00:05<00:08, 28.41it/s]
Loading 0:  20%|██        | 59/291 [00:05<00:06, 35.77it/s]
Loading 0:  23%|██▎       | 66/291 [00:06<00:07, 30.32it/s]
Loading 0:  24%|██▍       | 71/291 [00:06<00:06, 32.81it/s]
Loading 0:  27%|██▋       | 78/291 [00:06<00:06, 34.79it/s]
Loading 0:  30%|██▉       | 86/291 [00:06<00:04, 42.97it/s]
Loading 0:  32%|███▏      | 92/291 [00:06<00:04, 43.71it/s]
Loading 0:  34%|███▎      | 98/291 [00:06<00:04, 46.87it/s]
Loading 0:  36%|███▌      | 105/291 [00:06<00:04, 45.61it/s]
Loading 0:  39%|███▉      | 114/291 [00:07<00:03, 48.28it/s]
Loading 0:  42%|████▏     | 122/291 [00:07<00:03, 54.51it/s]
Loading 0:  44%|████▍     | 128/291 [00:07<00:03, 48.60it/s]
Loading 0:  46%|████▌     | 134/291 [00:07<00:03, 49.85it/s]
Loading 0:  48%|████▊     | 140/291 [00:07<00:02, 52.18it/s]
Loading 0:  50%|█████     | 146/291 [00:07<00:02, 49.16it/s]
Loading 0:  52%|█████▏    | 152/291 [00:07<00:02, 50.73it/s]
Loading 0:  55%|█████▍    | 159/291 [00:08<00:02, 47.52it/s]
Loading 0:  57%|█████▋    | 166/291 [00:08<00:03, 39.33it/s]
Loading 0:  59%|█████▉    | 172/291 [00:08<00:02, 43.28it/s]
Loading 0:  61%|██████    | 177/291 [00:08<00:02, 39.16it/s]
Loading 0:  64%|██████▎   | 185/291 [00:08<00:02, 47.73it/s]
Loading 0:  66%|██████▌   | 191/291 [00:08<00:02, 45.95it/s]
Loading 0:  67%|██████▋   | 196/291 [00:08<00:02, 46.37it/s]
Loading 0:  70%|██████▉   | 203/291 [00:08<00:01, 52.15it/s]
Loading 0:  72%|███████▏  | 209/291 [00:09<00:01, 49.84it/s]
Loading 0:  74%|███████▍  | 215/291 [00:09<00:01, 51.16it/s]
Loading 0:  76%|███████▌  | 221/291 [00:09<00:01, 53.12it/s]
Loading 0:  78%|███████▊  | 227/291 [00:09<00:01, 50.34it/s]
Loading 0:  80%|████████  | 233/291 [00:09<00:01, 52.33it/s]
Loading 0:  82%|████████▏ | 240/291 [00:09<00:01, 47.94it/s]
Loading 0:  85%|████████▌ | 248/291 [00:09<00:00, 55.21it/s]
Loading 0:  87%|████████▋ | 254/291 [00:09<00:00, 50.60it/s]
Loading 0:  89%|████████▉ | 260/291 [00:10<00:00, 51.60it/s]
Loading 0:  91%|█████████▏| 266/291 [00:10<00:00, 37.31it/s]
Loading 0:  93%|█████████▎| 272/291 [00:10<00:00, 38.56it/s]
Loading 0:  95%|█████████▌| 277/291 [00:10<00:00, 40.08it/s]
Loading 0:  98%|█████████▊| 284/291 [00:10<00:00, 46.07it/s]
Loading 0: 100%|█████████▉| 290/291 [00:10<00:00, 45.25it/s]
                                                            
/opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
pawankrd-cosmosrp-llama31-v1-mkmlizer:   warnings.warn(
pawankrd-cosmosrp-llama31-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
pawankrd-cosmosrp-llama31-v1-mkmlizer:   warnings.warn(
pawankrd-cosmosrp-llama31-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
pawankrd-cosmosrp-llama31-v1-mkmlizer: Saving duration: 1.426s
pawankrd-cosmosrp-llama31-v1-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.113s
pawankrd-cosmosrp-llama31-v1-mkmlizer: creating bucket guanaco-reward-models
pawankrd-cosmosrp-llama31-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
pawankrd-cosmosrp-llama31-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/config.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/special_tokens_map.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/tokenizer_config.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/merges.txt
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/vocab.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/tokenizer.json
pawankrd-cosmosrp-llama31-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/pawankrd-cosmosrp-llama31-v1_reward/reward.tensors
Job pawankrd-cosmosrp-llama31-v1-mkmlizer completed after 95.44s with status: succeeded
Stopping job with name pawankrd-cosmosrp-llama31-v1-mkmlizer
Pipeline stage MKMLizer completed in 96.60s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service pawankrd-cosmosrp-llama31-v1
Waiting for inference service pawankrd-cosmosrp-llama31-v1 to be ready
LLM-Router throws exception AssertionError('LLM-Router predict returns error 500') for galileo
LLM-Router throws exception AssertionError('LLM-Router predict returns error 500') for galileo
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 pawankrd-cosmosrp-llama31-v1 ready after 70.5766167640686s
Pipeline stage ISVCDeployer completed in 72.48s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2142515182495117s
Received healthy response to inference request in 1.4417531490325928s
Received healthy response to inference request in 1.3562347888946533s
Received healthy response to inference request in 1.3021655082702637s
Received healthy response to inference request in 1.5219025611877441s
5 requests
0 failed requests
5th percentile: 1.3129793643951415
10th percentile: 1.3237932205200196
20th percentile: 1.3454209327697755
30th percentile: 1.3733384609222412
40th percentile: 1.407545804977417
50th percentile: 1.4417531490325928
60th percentile: 1.4738129138946534
70th percentile: 1.5058726787567138
80th percentile: 1.6603723526000977
90th percentile: 1.9373119354248047
95th percentile: 2.075781726837158
99th percentile: 2.186557559967041
mean time: 1.5672615051269532
Pipeline stage StressChecker completed in 8.57s
pawankrd-cosmosrp-llama31_v1 status is now deployed due to DeploymentManager action
pawankrd-cosmosrp-llama31_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of pawankrd-cosmosrp-llama31_v1
Running pipeline stage ISVCDeleter
Checking if service pawankrd-cosmosrp-llama31-v1 is running
Tearing down inference service pawankrd-cosmosrp-llama31-v1
Service pawankrd-cosmosrp-llama31-v1 has been torndown
Pipeline stage ISVCDeleter completed in 4.64s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key pawankrd-cosmosrp-llama31-v1/config.json from bucket guanaco-mkml-models
Deleting key pawankrd-cosmosrp-llama31-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key pawankrd-cosmosrp-llama31-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key pawankrd-cosmosrp-llama31-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key pawankrd-cosmosrp-llama31-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key pawankrd-cosmosrp-llama31-v1_reward/config.json from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key pawankrd-cosmosrp-llama31-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.57s
pawankrd-cosmosrp-llama31_v1 status is now torndown due to DeploymentManager action