jic062-instruct

developer_uid: chace9580

submission_id: jic062-instruct_v7

model_name: jic062-instruct_v5

model_group: jic062/instruct

status: torndown

timestamp: 2024-07-29T05:42:26+00:00

num_battles: 12193

num_wins: 5952

celo_rating: 1192.16

family_friendly_score: 0.0

submission_type: basic

model_repo: jic062/instruct

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634

model_num_parameters: 8030269440.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: jic062-instruct_v5

is_internal_developer: False

language_model: jic062/instruct

model_size: 8B

ranking_group: single

us_pacific_date: 2024-07-28

win_ratio: 0.48814893791519726

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', '<|end_of_text|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}

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}

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'}

Resubmit model

Running pipeline stage MKMLizer
Starting job with name jic062-instruct-v7-mkmlizer
Waiting for job on jic062-instruct-v7-mkmlizer to finish
jic062-instruct-v7-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v7-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-v7-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-v7-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-v7-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-v7-mkmlizer: ║       /___/                                                         ║
jic062-instruct-v7-mkmlizer: ║                                                                     ║
jic062-instruct-v7-mkmlizer: ║  Version: 0.9.7                                                     ║
jic062-instruct-v7-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-v7-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-v7-mkmlizer: ║                                                                     ║
jic062-instruct-v7-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-v7-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-v7-mkmlizer: ║                                                                     ║
jic062-instruct-v7-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-v7-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-v7-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-v7-mkmlizer: ║                                                                     ║
jic062-instruct-v7-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission aetherwiing-mn-12b-starc_9812_v1: ('http://aetherwiing-mn-12b-starc-9812-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'request timeout')
jic062-instruct-v7-mkmlizer: Downloaded to shared memory in 21.185s
jic062-instruct-v7-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpgs50nver, device:0
jic062-instruct-v7-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v7-mkmlizer: quantized model in 25.527s
jic062-instruct-v7-mkmlizer: Processed model jic062/instruct in 46.713s
jic062-instruct-v7-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v7-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v7-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v7
jic062-instruct-v7-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v7/special_tokens_map.json
jic062-instruct-v7-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v7/config.json
jic062-instruct-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v7/tokenizer_config.json
jic062-instruct-v7-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v7/tokenizer.json
jic062-instruct-v7-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v7/flywheel_model.0.safetensors
jic062-instruct-v7-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v7-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 54.11it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:03, 70.69it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:02, 87.91it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:02, 85.96it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 84.15it/s]
Loading 0:  21%|██        | 61/291 [00:00<00:02, 81.56it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 89.52it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 24.15it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:06, 29.93it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:04, 38.14it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:03, 50.12it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:02, 55.62it/s]
Loading 0:  48%|████▊     | 141/291 [00:02<00:02, 65.22it/s]
Loading 0:  52%|█████▏    | 151/291 [00:02<00:02, 66.13it/s]
Loading 0:  55%|█████▍    | 160/291 [00:02<00:01, 70.50it/s]
Loading 0:  58%|█████▊    | 169/291 [00:03<00:01, 72.73it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 82.21it/s]
Loading 0:  66%|██████▋   | 193/291 [00:04<00:04, 24.24it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:03, 29.52it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 35.60it/s]
Loading 0:  76%|███████▋  | 222/291 [00:04<00:01, 45.60it/s]
Loading 0:  80%|███████▉  | 232/291 [00:04<00:01, 50.31it/s]
Loading 0:  83%|████████▎ | 241/291 [00:04<00:00, 56.82it/s]
Loading 0:  86%|████████▋ | 251/291 [00:04<00:00, 65.27it/s]
Loading 0:  91%|█████████ | 265/291 [00:05<00:00, 74.61it/s]
Loading 0:  94%|█████████▍| 274/291 [00:05<00:00, 77.94it/s]
Loading 0:  97%|█████████▋| 283/291 [00:05<00:00, 80.50it/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.
jic062-instruct-v7-mkmlizer:   warnings.warn(
jic062-instruct-v7-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jic062-instruct-v7-mkmlizer:   warnings.warn(
jic062-instruct-v7-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.
jic062-instruct-v7-mkmlizer:   warnings.warn(
jic062-instruct-v7-mkmlizer: 
Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]
Downloading shards:  50%|█████     | 1/2 [00:05<00:05,  5.10s/it]
Downloading shards: 100%|██████████| 2/2 [00:07<00:00,  3.74s/it]
Downloading shards: 100%|██████████| 2/2 [00:07<00:00,  3.94s/it]
jic062-instruct-v7-mkmlizer: 
Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]
Loading checkpoint shards:  50%|█████     | 1/2 [00:00<00:00,  2.46it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  4.09it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.72it/s]
jic062-instruct-v7-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v7-mkmlizer: Saving duration: 1.350s
jic062-instruct-v7-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 12.425s
jic062-instruct-v7-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v7-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v7-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v7_reward
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v7_reward/config.json
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v7_reward/special_tokens_map.json
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v7_reward/tokenizer_config.json
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v7_reward/merges.txt
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v7_reward/vocab.json
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v7_reward/tokenizer.json
jic062-instruct-v7-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-v7_reward/reward.tensors
Job jic062-instruct-v7-mkmlizer completed after 95.82s with status: succeeded
Stopping job with name jic062-instruct-v7-mkmlizer
Pipeline stage MKMLizer completed in 97.17s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v7
Waiting for inference service jic062-instruct-v7 to be ready
Inference service jic062-instruct-v7 ready after 101.09747862815857s
Pipeline stage ISVCDeployer completed in 102.44s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1801373958587646s
Received healthy response to inference request in 1.3804075717926025s
Received healthy response to inference request in 1.3861594200134277s
Received healthy response to inference request in 1.3449199199676514s
Received healthy response to inference request in 1.3747501373291016s
5 requests
0 failed requests
5th percentile: 1.3508859634399415
10th percentile: 1.3568520069122314
20th percentile: 1.3687840938568114
30th percentile: 1.3758816242218017
40th percentile: 1.3781445980072022
50th percentile: 1.3804075717926025
60th percentile: 1.3827083110809326
70th percentile: 1.3850090503692627
80th percentile: 1.5449550151824953
90th percentile: 1.8625462055206299
95th percentile: 2.0213418006896973
99th percentile: 2.148378276824951
mean time: 1.5332748889923096
Pipeline stage StressChecker completed in 8.30s
jic062-instruct_v7 status is now deployed due to DeploymentManager action
jic062-instruct_v7 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jic062-instruct_v7
Running pipeline stage ISVCDeleter
Checking if service jic062-instruct-v7 is running
Tearing down inference service jic062-instruct-v7
Service jic062-instruct-v7 has been torndown
Pipeline stage ISVCDeleter completed in 4.43s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jic062-instruct-v7/config.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v7/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jic062-instruct-v7/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v7/tokenizer.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v7/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jic062-instruct-v7_reward/config.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/merges.txt from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v7_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.46s
jic062-instruct_v7 status is now torndown due to DeploymentManager action