submission_id: jic062-instruct-v17_v1
developer_uid: chace9580
alignment_samples: 2057
alignment_score: 1.922095434274026
best_of: 16
celo_rating: 1226.42
display_name: jic062-instruct-v17_v1
formatter: {'memory_template': "<|begin_of_text|><|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': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_of_text|>', '|eot_id|'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: False
language_model: jic062/instruct_v17
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: jic062/instruct_v17
model_name: jic062-instruct-v17_v1
model_num_parameters: 8030261248.0
model_repo: jic062/instruct_v17
model_size: 8B
num_battles: 28680
num_wins: 14405
propriety_score: 0.7191426459719142
propriety_total_count: 1353.0
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: 2024-08-02T20:46:48+00:00
us_pacific_date: 2024-08-02
win_ratio: 0.5022663877266388
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jic062-instruct-v17-v1-mkmlizer
Waiting for job on jic062-instruct-v17-v1-mkmlizer to finish
jic062-instruct-v17-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v17-v1-mkmlizer: ║ _____ __ __ ║
jic062-instruct-v17-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-instruct-v17-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-instruct-v17-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-instruct-v17-v1-mkmlizer: ║ /___/ ║
jic062-instruct-v17-v1-mkmlizer: ║ ║
jic062-instruct-v17-v1-mkmlizer: ║ Version: 0.9.9 ║
jic062-instruct-v17-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-instruct-v17-v1-mkmlizer: ║ https://mk1.ai ║
jic062-instruct-v17-v1-mkmlizer: ║ ║
jic062-instruct-v17-v1-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-instruct-v17-v1-mkmlizer: ║ belonging to: ║
jic062-instruct-v17-v1-mkmlizer: ║ ║
jic062-instruct-v17-v1-mkmlizer: ║ Chai Research Corp. ║
jic062-instruct-v17-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-instruct-v17-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-instruct-v17-v1-mkmlizer: ║ ║
jic062-instruct-v17-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-v17-v1-mkmlizer: Downloaded to shared memory in 37.789s
jic062-instruct-v17-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpfbjkxr0j, device:0
jic062-instruct-v17-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v17-v1-mkmlizer: quantized model in 25.786s
jic062-instruct-v17-v1-mkmlizer: Processed model jic062/instruct_v17 in 63.576s
jic062-instruct-v17-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v17-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v17-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v17-v1
jic062-instruct-v17-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v17-v1/config.json
jic062-instruct-v17-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v17-v1/special_tokens_map.json
jic062-instruct-v17-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v17-v1/tokenizer_config.json
jic062-instruct-v17-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v17-v1/tokenizer.json
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
jic062-instruct-v17-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v17-v1/flywheel_model.0.safetensors
jic062-instruct-v17-v1-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v17-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 7/291 [00:00<00:05, 52.71it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:03, 71.69it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:03, 76.99it/s] Loading 0: 12%|█▏ | 34/291 [00:00<00:03, 79.83it/s] Loading 0: 15%|█▍ | 43/291 [00:00<00:02, 83.05it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:02, 83.87it/s] Loading 0: 21%|██ | 61/291 [00:00<00:02, 83.91it/s] Loading 0: 24%|██▍ | 70/291 [00:00<00:02, 84.83it/s] Loading 0: 27%|██▋ | 79/291 [00:00<00:02, 86.01it/s] Loading 0: 30%|███ | 88/291 [00:02<00:09, 21.28it/s] Loading 0: 34%|███▎ | 98/291 [00:02<00:06, 28.72it/s] Loading 0: 38%|███▊ | 112/291 [00:02<00:04, 39.85it/s] Loading 0: 42%|████▏ | 123/291 [00:02<00:03, 49.32it/s] Loading 0: 45%|████▌ | 132/291 [00:02<00:02, 55.92it/s] Loading 0: 49%|████▉ | 142/291 [00:02<00:02, 59.28it/s] Loading 0: 54%|█████▍ | 157/291 [00:02<00:01, 71.05it/s] Loading 0: 57%|█████▋ | 166/291 [00:02<00:01, 74.84it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:01, 77.35it/s] Loading 0: 63%|██████▎ | 184/291 [00:03<00:01, 75.27it/s] Loading 0: 66%|██████▋ | 193/291 [00:04<00:04, 22.27it/s] Loading 0: 69%|██████▉ | 202/291 [00:04<00:03, 28.19it/s] Loading 0: 73%|███████▎ | 211/291 [00:04<00:02, 35.16it/s] Loading 0: 77%|███████▋ | 223/291 [00:04<00:01, 44.10it/s] Loading 0: 80%|███████▉ | 232/291 [00:04<00:01, 51.29it/s] Loading 0: 83%|████████▎ | 241/291 [00:04<00:00, 57.60it/s] Loading 0: 86%|████████▌ | 250/291 [00:04<00:00, 63.68it/s] Loading 0: 89%|████████▉ | 260/291 [00:05<00:00, 71.69it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 79.50it/s] Loading 0: 98%|█████████▊| 284/291 [00:05<00:00, 83.21it/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-v17-v1-mkmlizer: warnings.warn(
jic062-instruct-v17-v1-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-v17-v1-mkmlizer: warnings.warn(
jic062-instruct-v17-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.
jic062-instruct-v17-v1-mkmlizer: warnings.warn(
jic062-instruct-v17-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v17-v1-mkmlizer: Saving duration: 1.379s
jic062-instruct-v17-v1-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.167s
jic062-instruct-v17-v1-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v17-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v17-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v17-v1_reward
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/special_tokens_map.json
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/tokenizer_config.json
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/config.json
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/merges.txt
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/vocab.json
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/tokenizer.json
jic062-instruct-v17-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-v17-v1_reward/reward.tensors
Job jic062-instruct-v17-v1-mkmlizer completed after 106.48s with status: succeeded
Stopping job with name jic062-instruct-v17-v1-mkmlizer
Pipeline stage MKMLizer completed in 107.65s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v17-v1
Waiting for inference service jic062-instruct-v17-v1 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 jic062-instruct-v17-v1 ready after 141.27833080291748s
Pipeline stage ISVCDeployer completed in 143.39s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1789939403533936s
Received healthy response to inference request in 1.4937663078308105s
Received healthy response to inference request in 1.445971965789795s
Received healthy response to inference request in 1.4362618923187256s
Received healthy response to inference request in 1.5108654499053955s
5 requests
0 failed requests
5th percentile: 1.4382039070129395
10th percentile: 1.4401459217071533
20th percentile: 1.444029951095581
30th percentile: 1.455530834197998
40th percentile: 1.4746485710144044
50th percentile: 1.4937663078308105
60th percentile: 1.5006059646606444
70th percentile: 1.5074456214904786
80th percentile: 1.6444911479949953
90th percentile: 1.9117425441741944
95th percentile: 2.0453682422637938
99th percentile: 2.1522688007354738
mean time: 1.6131719112396241
Pipeline stage StressChecker completed in 9.38s
jic062-instruct-v17_v1 status is now deployed due to DeploymentManager action
jic062-instruct-v17_v1 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct-v17_v1 status is now deployed due to admin request
jic062-instruct-v17_v1 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct-v17_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics