submission_id: hastagaras-llama-3-8b-ok_5775_v2
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/llama-3-8b-okay-two
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.9, 'top_p': 0.95, 'min_p': 0.0, 'top_k': 40, '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': "<|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: {'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}
timestamp: 2024-05-18T11:01:28+00:00
model_name: okay-2
model_eval_status: success
model_group: Hastagaras/llama-3-8b-ok
num_battles: 27064
num_wins: 14328
celo_rating: 1198.25
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: okay-2
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/llama-3-8b-okay-two
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-18
win_ratio: 0.5294117647058824
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama-3-8b-ok-5775-v2-mkmlizer
Waiting for job on hastagaras-llama-3-8b-ok-5775-v2-mkmlizer to finish
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ _____ __ __ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ /___/ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ https://mk1.ai ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ belonging to: ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ Chai Research Corp. ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama-3-8b-ok-5775-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-llama-3-8b-ok-5775-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Downloaded to shared memory in 16.450s
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 7%|▋ | 21/291 [00:00<00:01, 206.62it/s] Loading 0: 14%|█▍ | 42/291 [00:00<00:01, 196.88it/s] Loading 0: 23%|██▎ | 66/291 [00:00<00:01, 208.85it/s] Loading 0: 30%|██▉ | 87/291 [00:00<00:02, 98.89it/s] Loading 0: 38%|███▊ | 111/291 [00:00<00:01, 125.64it/s] Loading 0: 45%|████▌ | 131/291 [00:00<00:01, 139.44it/s] Loading 0: 54%|█████▎ | 156/291 [00:01<00:00, 161.75it/s] Loading 0: 62%|██████▏ | 179/291 [00:01<00:00, 175.16it/s] Loading 0: 68%|██████▊ | 199/291 [00:01<00:00, 106.22it/s] Loading 0: 76%|███████▌ | 220/291 [00:01<00:00, 122.34it/s] Loading 0: 82%|████████▏ | 239/291 [00:01<00:00, 135.52it/s] Loading 0: 91%|█████████ | 264/291 [00:01<00:00, 160.22it/s] Loading 0: 98%|█████████▊| 285/291 [00:01<00:00, 171.84it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: quantized model in 17.569s
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2/config.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2/special_tokens_map.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2/tokenizer_config.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2/tokenizer.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v2/flywheel_model.0.safetensors
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama-3-8b-ok-5775-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-llama-3-8b-ok-5775-v2-mkmlizer: warnings.warn(
hastagaras-llama-3-8b-ok-5775-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-llama-3-8b-ok-5775-v2-mkmlizer: warnings.warn(
hastagaras-llama-3-8b-ok-5775-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-llama-3-8b-ok-5775-v2-mkmlizer: warnings.warn(
hastagaras-llama-3-8b-ok-5775-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-llama-3-8b-ok-5775-v2-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Saving duration: 0.232s
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.265s
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/config.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/tokenizer_config.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/merges.txt
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/vocab.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/special_tokens_map.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/tokenizer.json
hastagaras-llama-3-8b-ok-5775-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v2_reward/reward.tensors
Job hastagaras-llama-3-8b-ok-5775-v2-mkmlizer completed after 62.85s with status: succeeded
Stopping job with name hastagaras-llama-3-8b-ok-5775-v2-mkmlizer
Pipeline stage MKMLizer completed in 65.93s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama-3-8b-ok-5775-v2
Waiting for inference service hastagaras-llama-3-8b-ok-5775-v2 to be ready
Inference service hastagaras-llama-3-8b-ok-5775-v2 ready after 40.20059895515442s
Pipeline stage ISVCDeployer completed in 47.07s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1059062480926514s
Received healthy response to inference request in 1.3933024406433105s
Received healthy response to inference request in 1.3401715755462646s
Received healthy response to inference request in 1.2918245792388916s
Received healthy response to inference request in 1.3751294612884521s
5 requests
0 failed requests
5th percentile: 1.3014939785003663
10th percentile: 1.3111633777618408
20th percentile: 1.33050217628479
30th percentile: 1.347163152694702
40th percentile: 1.3611463069915772
50th percentile: 1.3751294612884521
60th percentile: 1.3823986530303956
70th percentile: 1.389667844772339
80th percentile: 1.5358232021331788
90th percentile: 1.820864725112915
95th percentile: 1.9633854866027831
99th percentile: 2.0774020957946777
mean time: 1.501266860961914
Pipeline stage StressChecker completed in 8.20s
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 0.05s
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-llama-3-8b-ok_5775_v2 status is now deployed due to DeploymentManager action
hastagaras-llama-3-8b-ok_5775_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics