hastagaras-markites-llama3-8b

developer_uid: Hastagaras

submission_id: hastagaras-markites-llama3-8b_v2

model_name: mari-kita-tes

model_group: Hastagaras/Markites-llam

status: torndown

timestamp: 2024-05-16T14:58:43+00:00

num_battles: 8391

num_wins: 4382

celo_rating: 1189.78

family_friendly_score: 0.0

submission_type: basic

model_repo: Hastagaras/Markites-llama3-8b

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: mari-kita-tes

is_internal_developer: False

language_model: Hastagaras/Markites-llama3-8b

model_size: 8B

ranking_group: single

us_pacific_date: 2024-05-16

win_ratio: 0.5222261947324515

generation_params: {'temperature': 0.85, 'top_p': 0.95, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 1.05, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': '[[description]]\nBot name: {bot_name}\n{memory}\n\n', 'prompt_template': '{prompt}\n\n[[conversation]]', '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'}

model_eval_status: success

Resubmit model

Running pipeline stage MKMLizer
Starting job with name hastagaras-markites-llama3-8b-v2-mkmlizer
Waiting for job on hastagaras-markites-llama3-8b-v2-mkmlizer to finish
hastagaras-markites-llama3-8b-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-markites-llama3-8b-v2-mkmlizer: ║     _____            __           __                                ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║       /___/                                                         ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║                                                                     ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  Version: 0.8.14                                                    ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  https://mk1.ai                                                     ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║                                                                     ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  belonging to:                                                      ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║                                                                     ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ║                                                                     ║
hastagaras-markites-llama3-8b-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-markites-llama3-8b-v2-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-markites-llama3-8b-v2-mkmlizer:   warnings.warn(warning_message, FutureWarning)
hastagaras-markites-llama3-8b-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:01, 204.95it/s]
Loading 0:  14%|█▍        | 42/291 [00:00<00:01, 195.80it/s]
Loading 0:  23%|██▎       | 66/291 [00:00<00:01, 209.55it/s]
Loading 0:  30%|██▉       | 87/291 [00:00<00:02, 92.02it/s] 
Loading 0:  38%|███▊      | 111/291 [00:00<00:01, 117.65it/s]
Loading 0:  45%|████▌     | 131/291 [00:00<00:01, 132.07it/s]
Loading 0:  54%|█████▎    | 156/291 [00:01<00:00, 156.56it/s]
Loading 0:  62%|██████▏   | 179/291 [00:01<00:00, 171.27it/s]
Loading 0:  69%|██████▊   | 200/291 [00:01<00:00, 106.65it/s]
Loading 0:  76%|███████▌  | 220/291 [00:01<00:00, 120.86it/s]
Loading 0:  82%|████████▏ | 239/291 [00:01<00:00, 134.26it/s]
Loading 0:  91%|█████████ | 264/291 [00:01<00:00, 157.90it/s]
Loading 0:  98%|█████████▊| 284/291 [00:01<00:00, 167.33it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-markites-llama3-8b-v2-mkmlizer: quantized model in 17.561s
hastagaras-markites-llama3-8b-v2-mkmlizer: Processed model Hastagaras/Markites-llama3-8b in 35.493s
hastagaras-markites-llama3-8b-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-markites-llama3-8b-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-markites-llama3-8b-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2/special_tokens_map.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2/config.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2/tokenizer_config.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2/tokenizer.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-markites-llama3-8b-v2/flywheel_model.0.safetensors
hastagaras-markites-llama3-8b-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-markites-llama3-8b-v2-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-markites-llama3-8b-v2-mkmlizer:   warnings.warn(
hastagaras-markites-llama3-8b-v2-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-markites-llama3-8b-v2-mkmlizer:   warnings.warn(
hastagaras-markites-llama3-8b-v2-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-markites-llama3-8b-v2-mkmlizer:   warnings.warn(
hastagaras-markites-llama3-8b-v2-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-markites-llama3-8b-v2-mkmlizer:   return self.fget.__get__(instance, owner)()
hastagaras-markites-llama3-8b-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-markites-llama3-8b-v2-mkmlizer: Saving duration: 0.232s
hastagaras-markites-llama3-8b-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.205s
hastagaras-markites-llama3-8b-v2-mkmlizer: creating bucket guanaco-reward-models
hastagaras-markites-llama3-8b-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-markites-llama3-8b-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/special_tokens_map.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/config.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/merges.txt
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/tokenizer_config.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/vocab.json
hastagaras-markites-llama3-8b-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-markites-llama3-8b-v2_reward/tokenizer.json
Job hastagaras-markites-llama3-8b-v2-mkmlizer completed after 63.01s with status: succeeded
Stopping job with name hastagaras-markites-llama3-8b-v2-mkmlizer
Pipeline stage MKMLizer completed in 68.05s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-markites-llama3-8b-v2
Waiting for inference service hastagaras-markites-llama3-8b-v2 to be ready
Inference service hastagaras-markites-llama3-8b-v2 ready after 30.21074676513672s
Pipeline stage ISVCDeployer completed in 38.02s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.107478380203247s
Received healthy response to inference request in 1.2578768730163574s
Received healthy response to inference request in 1.2651004791259766s
Received healthy response to inference request in 1.2467381954193115s
Received healthy response to inference request in 1.2500886917114258s
5 requests
0 failed requests
5th percentile: 1.2474082946777343
10th percentile: 1.2480783939361573
20th percentile: 1.249418592453003
30th percentile: 1.2516463279724122
40th percentile: 1.2547616004943847
50th percentile: 1.2578768730163574
60th percentile: 1.2607663154602051
70th percentile: 1.2636557579040528
80th percentile: 1.4335760593414308
90th percentile: 1.770527219772339
95th percentile: 1.9390027999877928
99th percentile: 2.073783264160156
mean time: 1.4254565238952637
Pipeline stage StressChecker completed in 7.81s
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
hastagaras-markites-llama3-8b_v2 status is now deployed due to DeploymentManager action
hastagaras-markites-llama3-8b_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-markites-llama3-8b_v2
Running pipeline stage ISVCDeleter
Checking if service hastagaras-markites-llama3-8b-v2 is running
Tearing down inference service hastagaras-markites-llama3-8b-v2
Toredown service hastagaras-markites-llama3-8b-v2
Pipeline stage ISVCDeleter completed in 3.82s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-markites-llama3-8b-v2/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-markites-llama3-8b-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-markites-llama3-8b-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-markites-llama3-8b-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-markites-llama3-8b-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-markites-llama3-8b-v2_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-markites-llama3-8b-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.34s
hastagaras-markites-llama3-8b_v2 status is now torndown due to DeploymentManager action