submission_id: arlineka-llama3-rp_v2
developer_uid: arlineka
status: inactive
model_repo: arlineka/llama3-rp
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.1, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 100, '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: 2024-05-31T23:56:34+00:00
model_name: arlineka-llama3-rp_v1
model_eval_status: success
model_group: arlineka/llama3-rp
num_battles: 13160
num_wins: 7189
celo_rating: 1204.03
safety_score: 0.8
propriety_score: 1.0
propriety_total_count: 8.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: arlineka-llama3-rp_v1
ineligible_reason: propriety_total_count < 5000
language_model: arlineka/llama3-rp
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-31
win_ratio: 0.5462765957446809
Resubmit model
Running pipeline stage MKMLizer
Starting job with name arlineka-llama3-rp-v2-mkmlizer
Waiting for job on arlineka-llama3-rp-v2-mkmlizer to finish
arlineka-llama3-rp-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
arlineka-llama3-rp-v2-mkmlizer: ║ _____ __ __ ║
arlineka-llama3-rp-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
arlineka-llama3-rp-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
arlineka-llama3-rp-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
arlineka-llama3-rp-v2-mkmlizer: ║ /___/ ║
arlineka-llama3-rp-v2-mkmlizer: ║ ║
arlineka-llama3-rp-v2-mkmlizer: ║ Version: 0.8.14 ║
arlineka-llama3-rp-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
arlineka-llama3-rp-v2-mkmlizer: ║ https://mk1.ai ║
arlineka-llama3-rp-v2-mkmlizer: ║ ║
arlineka-llama3-rp-v2-mkmlizer: ║ The license key for the current software has been verified as ║
arlineka-llama3-rp-v2-mkmlizer: ║ belonging to: ║
arlineka-llama3-rp-v2-mkmlizer: ║ ║
arlineka-llama3-rp-v2-mkmlizer: ║ Chai Research Corp. ║
arlineka-llama3-rp-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
arlineka-llama3-rp-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
arlineka-llama3-rp-v2-mkmlizer: ║ ║
arlineka-llama3-rp-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
arlineka-llama3-rp-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.
arlineka-llama3-rp-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
arlineka-llama3-rp-v2-mkmlizer: Downloaded to shared memory in 27.864s
arlineka-llama3-rp-v2-mkmlizer: quantizing model to /dev/shm/model_cache
arlineka-llama3-rp-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
arlineka-llama3-rp-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:47, 2.24s/it] Loading 0: 5%|▍ | 14/291 [00:04<01:06, 4.14it/s] Loading 0: 11%|█ | 31/291 [00:04<00:23, 11.14it/s] Loading 0: 17%|█▋ | 49/291 [00:04<00:11, 20.58it/s] Loading 0: 22%|██▏ | 65/291 [00:04<00:07, 30.95it/s] Loading 0: 27%|██▋ | 80/291 [00:05<00:05, 42.16it/s] Loading 0: 33%|███▎ | 96/291 [00:05<00:03, 55.67it/s] Loading 0: 39%|███▉ | 114/291 [00:05<00:02, 72.38it/s] Loading 0: 45%|████▌ | 132/291 [00:05<00:01, 88.64it/s] Loading 0: 52%|█████▏ | 150/291 [00:05<00:01, 103.87it/s] Loading 0: 57%|█████▋ | 166/291 [00:06<00:02, 54.12it/s] Loading 0: 63%|██████▎ | 184/291 [00:06<00:01, 68.79it/s] Loading 0: 69%|██████▉ | 202/291 [00:06<00:01, 84.00it/s] Loading 0: 76%|███████▌ | 220/291 [00:06<00:00, 98.30it/s] Loading 0: 82%|████████▏ | 238/291 [00:06<00:00, 112.06it/s] Loading 0: 88%|████████▊ | 256/291 [00:06<00:00, 123.97it/s] Loading 0: 94%|█████████▍| 274/291 [00:06<00:00, 134.00it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
arlineka-llama3-rp-v2-mkmlizer: quantized model in 23.267s
arlineka-llama3-rp-v2-mkmlizer: Processed model arlineka/llama3-rp in 53.837s
arlineka-llama3-rp-v2-mkmlizer: creating bucket guanaco-mkml-models
arlineka-llama3-rp-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
arlineka-llama3-rp-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/arlineka-llama3-rp-v2
arlineka-llama3-rp-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/arlineka-llama3-rp-v2/special_tokens_map.json
arlineka-llama3-rp-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/arlineka-llama3-rp-v2/tokenizer.json
arlineka-llama3-rp-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/arlineka-llama3-rp-v2/config.json
arlineka-llama3-rp-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/arlineka-llama3-rp-v2/tokenizer_config.json
arlineka-llama3-rp-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/arlineka-llama3-rp-v2/flywheel_model.0.safetensors
arlineka-llama3-rp-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
arlineka-llama3-rp-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.
arlineka-llama3-rp-v2-mkmlizer: warnings.warn(
arlineka-llama3-rp-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.
arlineka-llama3-rp-v2-mkmlizer: warnings.warn(
arlineka-llama3-rp-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.
arlineka-llama3-rp-v2-mkmlizer: warnings.warn(
arlineka-llama3-rp-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()
arlineka-llama3-rp-v2-mkmlizer: return self.fget.__get__(instance, owner)()
arlineka-llama3-rp-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
arlineka-llama3-rp-v2-mkmlizer: Saving duration: 0.441s
arlineka-llama3-rp-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.422s
arlineka-llama3-rp-v2-mkmlizer: creating bucket guanaco-reward-models
arlineka-llama3-rp-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
arlineka-llama3-rp-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/special_tokens_map.json
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/config.json
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/merges.txt
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/vocab.json
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/tokenizer_config.json
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/tokenizer.json
arlineka-llama3-rp-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/arlineka-llama3-rp-v2_reward/reward.tensors
Job arlineka-llama3-rp-v2-mkmlizer completed after 83.96s with status: succeeded
Stopping job with name arlineka-llama3-rp-v2-mkmlizer
Pipeline stage MKMLizer completed in 87.98s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service arlineka-llama3-rp-v2
Waiting for inference service arlineka-llama3-rp-v2 to be ready
Inference service arlineka-llama3-rp-v2 ready after 30.187113285064697s
Pipeline stage ISVCDeployer completed in 37.23s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2185466289520264s
Received healthy response to inference request in 1.35921311378479s
Received healthy response to inference request in 1.3349568843841553s
Received healthy response to inference request in 1.324608325958252s
Received healthy response to inference request in 1.3627746105194092s
5 requests
0 failed requests
5th percentile: 1.3266780376434326
10th percentile: 1.3287477493286133
20th percentile: 1.3328871726989746
30th percentile: 1.3398081302642821
40th percentile: 1.349510622024536
50th percentile: 1.35921311378479
60th percentile: 1.3606377124786377
70th percentile: 1.3620623111724854
80th percentile: 1.5339290142059328
90th percentile: 1.8762378215789797
95th percentile: 2.0473922252655026
99th percentile: 2.1843157482147215
mean time: 1.5200199127197265
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 0.04s
M-Eval Dataset for topic stay_in_character is loaded
arlineka-llama3-rp_v2 status is now deployed due to DeploymentManager action
arlineka-llama3-rp_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics