developer_uid: v000000
submission_id: v000000-l3-8b-megaserpentine_v4
model_name: v000000-l3-8b-megaserpentine_v4
model_group: v000000/L3-8B-MegaSerpen
status: torndown
timestamp: 2024-06-18T07:43:27+00:00
num_battles: 20220
num_wins: 10436
celo_rating: 1209.45
family_friendly_score: 0.0
submission_type: basic
model_repo: v000000/L3-8B-MegaSerpentine
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: v000000-l3-8b-megaserpentine_v4
is_internal_developer: False
language_model: v000000/L3-8B-MegaSerpentine
model_size: 8B
ranking_group: single
us_pacific_date: 2024-06-18
win_ratio: 0.5161226508407517
generation_params: {'temperature': 1.3, 'top_p': 0.95, 'min_p': 0.08, 'top_k': 80, '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': "{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 v000000-l3-8b-megaserpentine-v4-mkmlizer
Waiting for job on v000000-l3-8b-megaserpentine-v4-mkmlizer to finish
v000000-l3-8b-megaserpentine-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║     _____            __           __                                ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║       /___/                                                         ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║                                                                     ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  Version: 0.8.14                                                    ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  https://mk1.ai                                                     ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║                                                                     ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  belonging to:                                                      ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║                                                                     ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  Chai Research Corp.                                                ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ║                                                                     ║
v000000-l3-8b-megaserpentine-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
v000000-l3-8b-megaserpentine-v4-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.
v000000-l3-8b-megaserpentine-v4-mkmlizer:   warnings.warn(warning_message, FutureWarning)
v000000-l3-8b-megaserpentine-v4-mkmlizer: Downloaded to shared memory in 17.856s
v000000-l3-8b-megaserpentine-v4-mkmlizer: quantizing model to /dev/shm/model_cache
v000000-l3-8b-megaserpentine-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
v000000-l3-8b-megaserpentine-v4-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<10:54,  2.27s/it]
Loading 0:   5%|▌         | 15/291 [00:04<01:02,  4.40it/s]
Loading 0:  11%|█▏        | 33/291 [00:04<00:22, 11.68it/s]
Loading 0:  18%|█▊        | 51/291 [00:04<00:11, 21.01it/s]
Loading 0:  22%|██▏       | 65/291 [00:05<00:09, 25.04it/s]
Loading 0:  27%|██▋       | 79/291 [00:05<00:06, 34.39it/s]
Loading 0:  33%|███▎      | 96/291 [00:05<00:04, 48.04it/s]
Loading 0:  39%|███▉      | 114/291 [00:05<00:02, 64.14it/s]
Loading 0:  45%|████▌     | 132/291 [00:05<00:01, 80.85it/s]
Loading 0:  52%|█████▏    | 150/291 [00:05<00:01, 96.23it/s]
Loading 0:  57%|█████▋    | 166/291 [00:06<00:01, 73.45it/s]
Loading 0:  63%|██████▎   | 182/291 [00:06<00:01, 87.29it/s]
Loading 0:  67%|██████▋   | 196/291 [00:06<00:00, 96.92it/s]
Loading 0:  73%|███████▎  | 213/291 [00:06<00:00, 110.97it/s]
Loading 0:  79%|███████▉  | 231/291 [00:06<00:00, 124.98it/s]
Loading 0:  86%|████████▌ | 249/291 [00:06<00:00, 133.59it/s]
Loading 0:  91%|█████████▏| 266/291 [00:06<00:00, 86.21it/s] 
Loading 0:  97%|█████████▋| 283/291 [00:07<00:00, 100.32it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
v000000-l3-8b-megaserpentine-v4-mkmlizer: quantized model in 23.429s
v000000-l3-8b-megaserpentine-v4-mkmlizer: Processed model v000000/L3-8B-MegaSerpentine in 43.963s
v000000-l3-8b-megaserpentine-v4-mkmlizer: creating bucket guanaco-mkml-models
v000000-l3-8b-megaserpentine-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
v000000-l3-8b-megaserpentine-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/v000000-l3-8b-megaserpentine-v4
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/v000000-l3-8b-megaserpentine-v4/config.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/v000000-l3-8b-megaserpentine-v4/tokenizer_config.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/v000000-l3-8b-megaserpentine-v4/special_tokens_map.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/v000000-l3-8b-megaserpentine-v4/tokenizer.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
v000000-l3-8b-megaserpentine-v4-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.
v000000-l3-8b-megaserpentine-v4-mkmlizer:   warnings.warn(
v000000-l3-8b-megaserpentine-v4-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.
v000000-l3-8b-megaserpentine-v4-mkmlizer:   warnings.warn(
v000000-l3-8b-megaserpentine-v4-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.
v000000-l3-8b-megaserpentine-v4-mkmlizer:   warnings.warn(
v000000-l3-8b-megaserpentine-v4-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()
v000000-l3-8b-megaserpentine-v4-mkmlizer:   return self.fget.__get__(instance, owner)()
v000000-l3-8b-megaserpentine-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
v000000-l3-8b-megaserpentine-v4-mkmlizer: Saving duration: 0.412s
v000000-l3-8b-megaserpentine-v4-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.764s
v000000-l3-8b-megaserpentine-v4-mkmlizer: creating bucket guanaco-reward-models
v000000-l3-8b-megaserpentine-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
v000000-l3-8b-megaserpentine-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/special_tokens_map.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/tokenizer_config.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/config.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/merges.txt
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/vocab.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/tokenizer.json
v000000-l3-8b-megaserpentine-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/v000000-l3-8b-megaserpentine-v4_reward/reward.tensors
Job v000000-l3-8b-megaserpentine-v4-mkmlizer completed after 72.74s with status: succeeded
Stopping job with name v000000-l3-8b-megaserpentine-v4-mkmlizer
Pipeline stage MKMLizer completed in 73.52s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service v000000-l3-8b-megaserpentine-v4
Waiting for inference service v000000-l3-8b-megaserpentine-v4 to be ready
Inference service v000000-l3-8b-megaserpentine-v4 ready after 110.67683172225952s
Pipeline stage ISVCDeployer completed in 116.47s
Running pipeline stage StressChecker
%s, retrying in %s seconds...
Received healthy response to inference request in 2.3183517456054688s
Received healthy response to inference request in 1.284388542175293s
Received healthy response to inference request in 1.2439086437225342s
Received healthy response to inference request in 1.2742772102355957s
%s, retrying in %s seconds...
Received healthy response to inference request in 1.3143126964569092s
Received healthy response to inference request in 1.275404691696167s
Received healthy response to inference request in 1.1367220878601074s
Received healthy response to inference request in 1.2510371208190918s
Received healthy response to inference request in 1.2682204246520996s
5 requests
0 failed requests
5th percentile: 1.1595850944519044
10th percentile: 1.1824481010437011
20th percentile: 1.2281741142272948
30th percentile: 1.2544737815856934
40th percentile: 1.2613471031188965
50th percentile: 1.2682204246520996
60th percentile: 1.2710941314697266
70th percentile: 1.2739678382873536
80th percentile: 1.2831862926483155
90th percentile: 1.2987494945526123
95th percentile: 1.3065310955047607
99th percentile: 1.3127563762664796
mean time: 1.249139404296875
Pipeline stage StressChecker completed in 53.24s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
v000000-l3-8b-megaserpentine_v4 status is now deployed due to DeploymentManager action
v000000-l3-8b-megaserpentine_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of v000000-l3-8b-megaserpentine_v4
Running pipeline stage ISVCDeleter
Checking if service v000000-l3-8b-megaserpentine-v4 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 3.30s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key v000000-l3-8b-megaserpentine-v4/config.json from bucket guanaco-mkml-models
Deleting key v000000-l3-8b-megaserpentine-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key v000000-l3-8b-megaserpentine-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key v000000-l3-8b-megaserpentine-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key v000000-l3-8b-megaserpentine-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key v000000-l3-8b-megaserpentine-v4_reward/config.json from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key v000000-l3-8b-megaserpentine-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.38s
v000000-l3-8b-megaserpentine_v4 status is now torndown due to DeploymentManager action