gryphe-pantheon-rp-8b-llama-3

developer_uid: Gryphe

submission_id: gryphe-pantheon-rp-8b-llama-3_v5

model_name: gryphe-pantheon-rp-8b-l3_v1

model_group: Gryphe/Pantheon-RP-8b-Ll

status: torndown

timestamp: 2024-05-15T16:13:25+00:00

num_battles: 17725

num_wins: 9563

celo_rating: 1201.28

family_friendly_score: 0.0

submission_type: basic

model_repo: Gryphe/Pantheon-RP-8b-Llama-3

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: gryphe-pantheon-rp-8b-l3_v1

is_internal_developer: False

language_model: Gryphe/Pantheon-RP-8b-Llama-3

model_size: 8B

ranking_group: single

us_pacific_date: 2024-05-15

win_ratio: 0.5395204513399153

generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 200, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "<|im_start|>system\n{bot_name}'s personality: {memory}\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}

model_eval_status: success

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

Resubmit model

Running pipeline stage MKMLizer
Starting job with name gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer
Waiting for job on gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer to finish
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║     _____            __           __                                ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║       /___/                                                         ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║                                                                     ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  Version: 0.8.14                                                    ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  https://mk1.ai                                                     ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║                                                                     ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  The license key for the current software has been verified as      ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  belonging to:                                                      ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║                                                                     ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  Chai Research Corp.                                                ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ║                                                                     ║
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer:   warnings.warn(warning_message, FutureWarning)
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Downloaded to shared memory in 174.512s
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: quantizing model to /dev/shm/model_cache
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:02, 127.11it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:01, 152.14it/s]
Loading 0:  17%|█▋        | 50/291 [00:00<00:01, 160.05it/s]
Loading 0:  23%|██▎       | 68/291 [00:00<00:01, 164.41it/s]
Loading 0:  29%|██▉       | 85/291 [00:00<00:02, 88.81it/s] 
Loading 0:  35%|███▌      | 103/291 [00:00<00:01, 107.10it/s]
Loading 0:  42%|████▏     | 121/291 [00:00<00:01, 122.32it/s]
Loading 0:  48%|████▊     | 139/291 [00:01<00:01, 134.76it/s]
Loading 0:  54%|█████▍    | 157/291 [00:01<00:00, 145.08it/s]
Loading 0:  61%|██████    | 177/291 [00:01<00:00, 159.27it/s]
Loading 0:  67%|██████▋   | 195/291 [00:01<00:01, 95.28it/s] 
Loading 0:  73%|███████▎  | 212/291 [00:01<00:00, 108.46it/s]
Loading 0:  79%|███████▉  | 230/291 [00:01<00:00, 122.75it/s]
Loading 0:  85%|████████▌ | 248/291 [00:01<00:00, 135.44it/s]
Loading 0:  91%|█████████▏| 266/291 [00:02<00:00, 145.52it/s]
Loading 0:  98%|█████████▊| 286/291 [00:02<00:00, 159.24it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: quantized model in 22.762s
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Processed model Gryphe/Pantheon-RP-8b-Llama-3 in 199.751s
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: creating bucket guanaco-mkml-models
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5/config.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5/special_tokens_map.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5/tokenizer_config.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5/tokenizer.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/gryphe-pantheon-rp-8b-llama-3-v5/flywheel_model.0.safetensors
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer:   warnings.warn(
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer:   warnings.warn(
gryphe-pantheon-rp-8b-llama-3-v5-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.
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer:   warnings.warn(
gryphe-pantheon-rp-8b-llama-3-v5-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()
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer:   return self.fget.__get__(instance, owner)()
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Saving duration: 0.407s
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 17.971s
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: creating bucket guanaco-reward-models
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: Bucket 's3://guanaco-reward-models/' created
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/config.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/special_tokens_map.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/tokenizer_config.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/merges.txt
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/vocab.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/tokenizer.json
gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/gryphe-pantheon-rp-8b-llama-3-v5_reward/reward.tensors
Job gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer completed after 295.25s with status: succeeded
Stopping job with name gryphe-pantheon-rp-8b-llama-3-v5-mkmlizer
Pipeline stage MKMLizer completed in 300.18s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service gryphe-pantheon-rp-8b-llama-3-v5
Waiting for inference service gryphe-pantheon-rp-8b-llama-3-v5 to be ready
Inference service gryphe-pantheon-rp-8b-llama-3-v5 ready after 40.33631491661072s
Pipeline stage ISVCDeployer completed in 46.81s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.203479766845703s
Received healthy response to inference request in 1.3116066455841064s
Received healthy response to inference request in 1.2954192161560059s
Received healthy response to inference request in 1.2906436920166016s
Received healthy response to inference request in 1.2982499599456787s
5 requests
0 failed requests
5th percentile: 1.2915987968444824
10th percentile: 1.2925539016723633
20th percentile: 1.294464111328125
30th percentile: 1.2959853649139403
40th percentile: 1.2971176624298095
50th percentile: 1.2982499599456787
60th percentile: 1.3035926342010498
70th percentile: 1.3089353084564208
80th percentile: 1.4899812698364259
90th percentile: 1.8467305183410645
95th percentile: 2.0251051425933837
99th percentile: 2.167804841995239
mean time: 1.4798798561096191
Pipeline stage StressChecker completed in 8.02s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running M-Eval for topic stay_in_character
Running pipeline stage DaemonicSafetyScorer
M-Eval Dataset for topic stay_in_character is loaded
Pipeline stage DaemonicSafetyScorer completed in 0.07s
gryphe-pantheon-rp-8b-llama-3_v5 status is now deployed due to DeploymentManager action
gryphe-pantheon-rp-8b-llama-3_v5 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of gryphe-pantheon-rp-8b-llama-3_v5
Running pipeline stage ISVCDeleter
Checking if service gryphe-pantheon-rp-8b-llama-3-v5 is running
Tearing down inference service gryphe-pantheon-rp-8b-llama-3-v5
Toredown service gryphe-pantheon-rp-8b-llama-3-v5
Pipeline stage ISVCDeleter completed in 3.28s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key gryphe-pantheon-rp-8b-llama-3-v5/config.json from bucket guanaco-mkml-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5/tokenizer.json from bucket guanaco-mkml-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/config.json from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/merges.txt from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/reward.tensors from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key gryphe-pantheon-rp-8b-llama-3-v5_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.61s
gryphe-pantheon-rp-8b-llama-3_v5 status is now torndown due to DeploymentManager action