jellywibble-nguy-alignme_4960

submission_id: jellywibble-nguy-alignme_4960_v1

developer_uid: Jellywibble

best_of: 16

celo_rating: 1151.26

display_name: meseca-20062024-c1_v1

family_friendly_score: 0.0

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': 0.95, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 60, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

is_internal_developer: True

language_model: Jellywibble/nguy-alignment-regression-cp128

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: Jellywibble/nguy-alignme

model_name: meseca-20062024-c1_v1

model_num_parameters: 8030261248.0

model_repo: Jellywibble/nguy-alignment-regression-cp128

model_size: 8B

num_battles: 31528

num_wins: 14527

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/reward_gpt2_medium_preference_24m_e2

status: torndown

submission_type: basic

timestamp: 2024-07-05T20:09:01+00:00

us_pacific_date: 2024-07-05

win_ratio: 0.46076503425526516

Resubmit model

Running pipeline stage MKMLizer
Starting job with name jellywibble-nguy-alignme-4960-v1-mkmlizer
Waiting for job on jellywibble-nguy-alignme-4960-v1-mkmlizer to finish
jellywibble-nguy-alignme-4960-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║     _____            __           __                                ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║       /___/                                                         ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║                                                                     ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║                                                                     ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  belonging to:                                                      ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║                                                                     ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ║                                                                     ║
jellywibble-nguy-alignme-4960-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jellywibble-nguy-alignme-4960-v1-mkmlizer: Downloaded to shared memory in 35.062s
jellywibble-nguy-alignme-4960-v1-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-nguy-alignme-4960-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jellywibble-nguy-alignme-4960-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:02, 119.42it/s]
Loading 0:   8%|▊         | 24/291 [00:00<00:02, 108.31it/s]
Loading 0:  13%|█▎        | 39/291 [00:00<00:02, 120.79it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 113.75it/s]
Loading 0:  23%|██▎       | 66/291 [00:00<00:01, 117.06it/s]
Loading 0:  27%|██▋       | 78/291 [00:00<00:01, 111.36it/s]
Loading 0:  31%|███       | 90/291 [00:01<00:03, 62.53it/s] 
Loading 0:  35%|███▌      | 103/291 [00:01<00:02, 72.86it/s]
Loading 0:  40%|███▉      | 116/291 [00:01<00:02, 84.00it/s]
Loading 0:  44%|████▍     | 129/291 [00:01<00:01, 93.89it/s]
Loading 0:  48%|████▊     | 141/291 [00:01<00:01, 95.70it/s]
Loading 0:  54%|█████▎    | 156/291 [00:01<00:01, 106.46it/s]
Loading 0:  58%|█████▊    | 168/291 [00:01<00:01, 104.64it/s]
Loading 0:  62%|██████▏   | 181/291 [00:01<00:01, 107.34it/s]
Loading 0:  66%|██████▋   | 193/291 [00:02<00:01, 66.76it/s] 
Loading 0:  70%|██████▉   | 203/291 [00:02<00:01, 72.77it/s]
Loading 0:  75%|███████▌  | 219/291 [00:02<00:00, 87.86it/s]
Loading 0:  79%|███████▉  | 230/291 [00:02<00:00, 88.84it/s]
Loading 0:  85%|████████▍ | 246/291 [00:02<00:00, 102.60it/s]
Loading 0:  89%|████████▊ | 258/291 [00:02<00:00, 103.04it/s]
Loading 0:  94%|█████████▍| 273/291 [00:02<00:00, 113.09it/s]
Loading 0:  98%|█████████▊| 285/291 [00:02<00:00, 111.23it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
jellywibble-nguy-alignme-4960-v1-mkmlizer: quantized model in 24.461s
jellywibble-nguy-alignme-4960-v1-mkmlizer: Processed model Jellywibble/nguy-alignment-regression-cp128 in 59.523s
jellywibble-nguy-alignme-4960-v1-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-nguy-alignme-4960-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-nguy-alignme-4960-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1/special_tokens_map.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1/tokenizer_config.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1/config.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1/tokenizer.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-nguy-alignme-4960-v1/flywheel_model.0.safetensors
jellywibble-nguy-alignme-4960-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
jellywibble-nguy-alignme-4960-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-nguy-alignme-4960-v1-mkmlizer:   warnings.warn(
jellywibble-nguy-alignme-4960-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
jellywibble-nguy-alignme-4960-v1-mkmlizer:   warnings.warn(
jellywibble-nguy-alignme-4960-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-nguy-alignme-4960-v1-mkmlizer:   warnings.warn(
jellywibble-nguy-alignme-4960-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-nguy-alignme-4960-v1-mkmlizer:   warnings.warn(
jellywibble-nguy-alignme-4960-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
jellywibble-nguy-alignme-4960-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
jellywibble-nguy-alignme-4960-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jellywibble-nguy-alignme-4960-v1-mkmlizer: Saving duration: 0.403s
jellywibble-nguy-alignme-4960-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.534s
jellywibble-nguy-alignme-4960-v1-mkmlizer: creating bucket guanaco-reward-models
jellywibble-nguy-alignme-4960-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jellywibble-nguy-alignme-4960-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/config.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/merges.txt
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/tokenizer_config.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/special_tokens_map.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/tokenizer.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/vocab.json
jellywibble-nguy-alignme-4960-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jellywibble-nguy-alignme-4960-v1_reward/reward.tensors
Job jellywibble-nguy-alignme-4960-v1-mkmlizer completed after 83.79s with status: succeeded
Stopping job with name jellywibble-nguy-alignme-4960-v1-mkmlizer
Pipeline stage MKMLizer completed in 84.79s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.41s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-nguy-alignme-4960-v1
Waiting for inference service jellywibble-nguy-alignme-4960-v1 to be ready
Failed to get response for submission mistralai-mixtral-8x7b-_3473_v61: ('http://mistralai-mixtral-8x7b-3473-v61-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'upstream connect error or disconnect/reset before headers. reset reason: connection timeout')
Inference service jellywibble-nguy-alignme-4960-v1 ready after 50.272507429122925s
Pipeline stage ISVCDeployer completed in 57.22s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2960097789764404s
Received healthy response to inference request in 1.3245301246643066s
Received healthy response to inference request in 1.29927396774292s
Received healthy response to inference request in 1.2708101272583008s
Received healthy response to inference request in 1.332183599472046s
5 requests
0 failed requests
5th percentile: 1.2765028953552247
10th percentile: 1.2821956634521485
20th percentile: 1.293581199645996
30th percentile: 1.3043251991271974
40th percentile: 1.314427661895752
50th percentile: 1.3245301246643066
60th percentile: 1.3275915145874024
70th percentile: 1.330652904510498
80th percentile: 1.524948835372925
90th percentile: 1.9104793071746826
95th percentile: 2.1032445430755615
99th percentile: 2.257456731796265
mean time: 1.5045615196228028
Pipeline stage StressChecker completed in 8.44s
jellywibble-nguy-alignme_4960_v1 status is now deployed due to DeploymentManager action
jellywibble-nguy-alignme_4960_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jellywibble-nguy-alignme_4960_v1
Running pipeline stage ISVCDeleter
Checking if service jellywibble-nguy-alignme-4960-v1 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.16s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jellywibble-nguy-alignme-4960-v1/config.json from bucket guanaco-mkml-models
Deleting key jellywibble-nguy-alignme-4960-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jellywibble-nguy-alignme-4960-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jellywibble-nguy-alignme-4960-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key jellywibble-nguy-alignme-4960-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jellywibble-nguy-alignme-4960-v1_reward/config.json from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jellywibble-nguy-alignme-4960-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.36s
jellywibble-nguy-alignme_4960_v1 status is now torndown due to DeploymentManager action