hastagaras-sciemet-8b-l3_4154

developer_uid: Hastagaras

submission_id: hastagaras-sciemet-8b-l3_4154_v1

model_name: t

model_group: Hastagaras/Sciemet-8B-L3

status: torndown

timestamp: 2024-06-16T18:27:16+00:00

num_battles: 31016

num_wins: 17505

celo_rating: 1213.83

family_friendly_score: 0.0

submission_type: basic

model_repo: Hastagaras/Sciemet-8B-L3-MK.II-Evil

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: t

is_internal_developer: False

language_model: Hastagaras/Sciemet-8B-L3-MK.II-Evil

model_size: 8B

ranking_group: single

us_pacific_date: 2024-06-16

win_ratio: 0.5643861232912045

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 35, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "<|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}

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'}

model_eval_status: success

Resubmit model

Running pipeline stage MKMLizer
Starting job with name hastagaras-sciemet-8b-l3-4154-v1-mkmlizer
Waiting for job on hastagaras-sciemet-8b-l3-4154-v1-mkmlizer to finish
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║     _____            __           __                                ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║       /___/                                                         ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║                                                                     ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  https://mk1.ai                                                     ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║                                                                     ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  belonging to:                                                      ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║                                                                     ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ║                                                                     ║
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-sciemet-8b-l3-4154-v1-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.
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Downloaded to shared memory in 25.288s
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:02, 123.72it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:01, 150.88it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:01, 157.32it/s]
Loading 0:  22%|██▏       | 65/291 [00:00<00:01, 154.50it/s]
Loading 0:  28%|██▊       | 81/291 [00:00<00:01, 151.82it/s]
Loading 0:  33%|███▎      | 97/291 [00:00<00:02, 82.69it/s] 
Loading 0:  39%|███▉      | 113/291 [00:01<00:01, 96.40it/s]
Loading 0:  45%|████▌     | 131/291 [00:01<00:01, 110.95it/s]
Loading 0:  51%|█████     | 148/291 [00:01<00:01, 123.22it/s]
Loading 0:  57%|█████▋    | 165/291 [00:01<00:00, 132.15it/s]
Loading 0:  62%|██████▏   | 181/291 [00:01<00:00, 135.83it/s]
Loading 0:  67%|██████▋   | 196/291 [00:01<00:01, 86.75it/s] 
Loading 0:  73%|███████▎  | 212/291 [00:01<00:00, 99.73it/s]
Loading 0:  79%|███████▉  | 230/291 [00:01<00:00, 115.13it/s]
Loading 0:  85%|████████▍ | 247/291 [00:02<00:00, 127.71it/s]
Loading 0:  91%|█████████ | 265/291 [00:02<00:00, 136.23it/s]
Loading 0:  97%|█████████▋| 282/291 [00:02<00:00, 142.55it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: quantized model in 22.694s
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Processed model Hastagaras/Sciemet-8B-L3-MK.II-Evil in 50.448s
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1/tokenizer_config.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1/config.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1/special_tokens_map.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1/tokenizer.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-sciemet-8b-l3-4154-v1/flywheel_model.0.safetensors
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-sciemet-8b-l3-4154-v1-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.
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer:   warnings.warn(
hastagaras-sciemet-8b-l3-4154-v1-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.
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer:   warnings.warn(
hastagaras-sciemet-8b-l3-4154-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.
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer:   warnings.warn(
hastagaras-sciemet-8b-l3-4154-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()
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Saving duration: 0.402s
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.485s
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/special_tokens_map.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/config.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/tokenizer.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/reward.tensors
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/tokenizer_config.json
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/merges.txt
hastagaras-sciemet-8b-l3-4154-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-sciemet-8b-l3-4154-v1_reward/vocab.json
Job hastagaras-sciemet-8b-l3-4154-v1-mkmlizer completed after 103.71s with status: succeeded
Stopping job with name hastagaras-sciemet-8b-l3-4154-v1-mkmlizer
Pipeline stage MKMLizer completed in 106.53s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-sciemet-8b-l3-4154-v1
Waiting for inference service hastagaras-sciemet-8b-l3-4154-v1 to be ready
Inference service hastagaras-sciemet-8b-l3-4154-v1 ready after 40.21808910369873s
Pipeline stage ISVCDeployer completed in 46.88s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.203672409057617s
Received healthy response to inference request in 16.772868394851685s
Received healthy response to inference request in 1.3466389179229736s
Received healthy response to inference request in 19.183006048202515s
Received healthy response to inference request in 1.2489433288574219s
5 requests
0 failed requests
5th percentile: 1.2684824466705322
10th percentile: 1.2880215644836426
20th percentile: 1.3270998001098633
30th percentile: 1.5180456161499023
40th percentile: 1.8608590126037599
50th percentile: 2.203672409057617
60th percentile: 8.031350803375243
70th percentile: 13.859029197692868
80th percentile: 17.25489592552185
90th percentile: 18.218950986862183
95th percentile: 18.70097851753235
99th percentile: 19.08660054206848
mean time: 8.151025819778443
%s, retrying in %s seconds...
Received healthy response to inference request in 1.2552177906036377s
Received healthy response to inference request in 1.380075454711914s
Received healthy response to inference request in 1.324737548828125s
Received healthy response to inference request in 1.303558588027954s
Received healthy response to inference request in 10.364341974258423s
5 requests
0 failed requests
5th percentile: 1.264885950088501
10th percentile: 1.2745541095733643
20th percentile: 1.2938904285430908
30th percentile: 1.3077943801879883
40th percentile: 1.3162659645080566
50th percentile: 1.324737548828125
60th percentile: 1.3468727111816405
70th percentile: 1.3690078735351563
80th percentile: 3.1769287586212176
90th percentile: 6.77063536643982
95th percentile: 8.56748867034912
99th percentile: 10.004971313476561
mean time: 3.1255862712860107
Pipeline stage StressChecker completed in 57.66s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-sciemet-8b-l3_4154_v1 status is now deployed due to DeploymentManager action
hastagaras-sciemet-8b-l3_4154_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-sciemet-8b-l3_4154_v1
Running pipeline stage ISVCDeleter
Checking if service hastagaras-sciemet-8b-l3-4154-v1 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.31s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-sciemet-8b-l3-4154-v1/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-sciemet-8b-l3-4154-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 6.26s
hastagaras-sciemet-8b-l3_4154_v1 status is now torndown due to DeploymentManager action