submission_id: jic062-instruct-v14-v15_v3
developer_uid: chace9580
alignment_samples: 11729
alignment_score: 0.7744277940211114
best_of: 16
celo_rating: 1226.07
display_name: jic062-instruct_resubmit_rwch
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}
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_v14_v15
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: jic062/instruct_v14_v15
model_name: jic062-instruct_resubmit_rwch
model_num_parameters: 8030261248.0
model_repo: jic062/instruct_v14_v15
model_size: 8B
num_battles: 11729
num_wins: 5885
propriety_score: 0.7055393586005831
propriety_total_count: 1029.0
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: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: basic
timestamp: 2024-08-12T07:32:33+00:00
us_pacific_date: 2024-08-12
win_ratio: 0.5017478045869213
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jic062-instruct-v14-v15-v3-mkmlizer
Waiting for job on jic062-instruct-v14-v15-v3-mkmlizer to finish
Stopping job with name jic062-instruct-v14-v15-v3-mkmlizer
%s, retrying in %s seconds...
Starting job with name jic062-instruct-v14-v15-v3-mkmlizer
Waiting for job on jic062-instruct-v14-v15-v3-mkmlizer to finish
jic062-instruct-v14-v15-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-v14-v15-v3-mkmlizer: ║ _____ __ __ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ /___/ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ Version: 0.9.9 ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ https://mk1.ai ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ belonging to: ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ Chai Research Corp. ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-instruct-v14-v15-v3-mkmlizer: ║ ║
jic062-instruct-v14-v15-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-v14-v15-v3-mkmlizer: Downloaded to shared memory in 22.726s
jic062-instruct-v14-v15-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpv7nyx4sk, device:0
jic062-instruct-v14-v15-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-v14-v15-v3-mkmlizer: quantized model in 25.883s
jic062-instruct-v14-v15-v3-mkmlizer: Processed model jic062/instruct_v14_v15 in 48.609s
jic062-instruct-v14-v15-v3-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-v14-v15-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-v14-v15-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3
jic062-instruct-v14-v15-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3/special_tokens_map.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3/config.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3/tokenizer_config.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3/tokenizer.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-instruct-v14-v15-v3/flywheel_model.0.safetensors
jic062-instruct-v14-v15-v3-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-v14-v15-v3-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 7/291 [00:00<00:05, 49.22it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:03, 69.35it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:03, 77.68it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:02, 89.66it/s] Loading 0: 17%|█▋ | 49/291 [00:00<00:02, 88.26it/s] Loading 0: 20%|█▉ | 58/291 [00:00<00:02, 86.63it/s] Loading 0: 24%|██▍ | 70/291 [00:00<00:02, 86.31it/s] Loading 0: 29%|██▊ | 83/291 [00:02<00:08, 24.82it/s] Loading 0: 31%|███ | 90/291 [00:02<00:07, 28.59it/s] Loading 0: 35%|███▌ | 103/291 [00:02<00:04, 37.76it/s] Loading 0: 38%|███▊ | 112/291 [00:02<00:04, 44.49it/s] Loading 0: 42%|████▏ | 121/291 [00:02<00:03, 50.34it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:02, 57.01it/s] Loading 0: 48%|████▊ | 139/291 [00:02<00:02, 63.58it/s] Loading 0: 51%|█████ | 148/291 [00:02<00:02, 69.30it/s] Loading 0: 54%|█████▍ | 157/291 [00:02<00:01, 73.86it/s] Loading 0: 58%|█████▊ | 169/291 [00:03<00:01, 77.27it/s] Loading 0: 62%|██████▏ | 179/291 [00:03<00:01, 82.27it/s] Loading 0: 65%|██████▍ | 188/291 [00:04<00:04, 22.75it/s] Loading 0: 68%|██████▊ | 197/291 [00:04<00:03, 28.85it/s] Loading 0: 70%|███████ | 205/291 [00:04<00:02, 33.89it/s] Loading 0: 76%|███████▌ | 220/291 [00:04<00:01, 47.33it/s] Loading 0: 80%|███████▉ | 232/291 [00:04<00:01, 54.82it/s] Loading 0: 85%|████████▍ | 247/291 [00:04<00:00, 66.70it/s] Loading 0: 89%|████████▉ | 259/291 [00:05<00:00, 71.29it/s] Loading 0: 94%|█████████▍| 274/291 [00:05<00:00, 80.50it/s] Loading 0: 98%|█████████▊| 286/291 [00:05<00:00, 81.41it/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-v14-v15-v3-mkmlizer: warnings.warn(
jic062-instruct-v14-v15-v3-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-v14-v15-v3-mkmlizer: warnings.warn(
jic062-instruct-v14-v15-v3-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-v14-v15-v3-mkmlizer: warnings.warn(
jic062-instruct-v14-v15-v3-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:05<00:05, 5.52s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.49s/it] Downloading shards: 100%|██████████| 2/2 [00:09<00:00, 4.64s/it]
jic062-instruct-v14-v15-v3-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 2.45it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 4.03it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 3.67it/s]
jic062-instruct-v14-v15-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-v14-v15-v3-mkmlizer: Saving duration: 1.391s
jic062-instruct-v14-v15-v3-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 14.501s
jic062-instruct-v14-v15-v3-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-v14-v15-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-v14-v15-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/config.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/special_tokens_map.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/tokenizer_config.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/merges.txt
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/vocab.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/tokenizer.json
jic062-instruct-v14-v15-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-v14-v15-v3_reward/reward.tensors
Job jic062-instruct-v14-v15-v3-mkmlizer completed after 94.33s with status: succeeded
Stopping job with name jic062-instruct-v14-v15-v3-mkmlizer
Pipeline stage MKMLizer completed in 96.08s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-v14-v15-v3
Waiting for inference service jic062-instruct-v14-v15-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
Failed to get response for submission mistralai-mistral-nemo-_9330_v42: ('http://mistralai-mistral-nemo-9330-v42-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59228->127.0.0.1:8080: read: connection reset by peer\n')
Inference service jic062-instruct-v14-v15-v3 ready after 201.2468183040619s
Pipeline stage ISVCDeployer completed in 202.98s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.229616641998291s
Received healthy response to inference request in 1.392087459564209s
Received healthy response to inference request in 1.4004735946655273s
Received healthy response to inference request in 1.4886760711669922s
Received healthy response to inference request in 1.4150705337524414s
5 requests
0 failed requests
5th percentile: 1.3937646865844726
10th percentile: 1.3954419136047362
20th percentile: 1.3987963676452637
30th percentile: 1.4033929824829101
40th percentile: 1.4092317581176759
50th percentile: 1.4150705337524414
60th percentile: 1.4445127487182616
70th percentile: 1.473954963684082
80th percentile: 1.6368641853332522
90th percentile: 1.9332404136657715
95th percentile: 2.0814285278320312
99th percentile: 2.1999790191650392
mean time: 1.5851848602294922
Pipeline stage StressChecker completed in 8.88s
jic062-instruct-v14-v15_v3 status is now deployed due to DeploymentManager action
jic062-instruct-v14-v15_v3 status is now inactive due to auto deactivation removed underperforming models
jic062-instruct-v14-v15_v3 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics