submission_id: hastagaras-llama-3-8b-ok_5775_v6
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': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 200, '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 personality: {memory}\n\n", 'prompt_template': 'Scenario: {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: 2024-05-20T13:15:16+00:00
model_name: new-okay-two-repetition-fixed
model_eval_status: success
model_group: Hastagaras/llama-3-8b-ok
num_battles: 25849
num_wins: 14041
celo_rating: 1198.9
safety_score: 0.91
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: new-okay-two-repetition-fixed
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: 2024-05-20
win_ratio: 0.5431931602769933
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama-3-8b-ok-5775-v6-mkmlizer
Waiting for job on hastagaras-llama-3-8b-ok-5775-v6-mkmlizer to finish
Stopping job with name hastagaras-llama-3-8b-ok-5775-v6-mkmlizer
%s, retrying in %s seconds...
Starting job with name hastagaras-llama-3-8b-ok-5775-v6-mkmlizer
Waiting for job on hastagaras-llama-3-8b-ok-5775-v6-mkmlizer to finish
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ _____ __ __ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ /___/ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ https://mk1.ai ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ belonging to: ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ Chai Research Corp. ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ║ ║
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama-3-8b-ok-5775-v6-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-v6-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Downloaded to shared memory in 13.248s
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 5%|▍ | 14/291 [00:00<00:02, 124.81it/s] Loading 0: 11%|█ | 32/291 [00:00<00:01, 146.71it/s] Loading 0: 17%|█▋ | 50/291 [00:00<00:01, 151.63it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 156.16it/s] Loading 0: 29%|██▊ | 83/291 [00:00<00:02, 76.69it/s] Loading 0: 33%|███▎ | 96/291 [00:00<00:02, 86.86it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:01, 102.53it/s] Loading 0: 45%|████▌ | 131/291 [00:01<00:01, 117.91it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 129.20it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:00, 138.58it/s] Loading 0: 64%|██████▍ | 187/291 [00:01<00:01, 92.32it/s] Loading 0: 69%|██████▉ | 201/291 [00:01<00:00, 101.00it/s] Loading 0: 75%|███████▌ | 219/291 [00:01<00:00, 115.18it/s] Loading 0: 81%|████████▏ | 237/291 [00:02<00:00, 127.53it/s] Loading 0: 88%|████████▊ | 255/291 [00:02<00:00, 137.06it/s] Loading 0: 94%|█████████▍| 273/291 [00:02<00:00, 143.36it/s] Loading 0: 99%|█████████▉| 289/291 [00:07<00:00, 10.35it/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-v6-mkmlizer: quantized model in 18.518s
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Processed model Hastagaras/llama-3-8b-okay-two in 32.790s
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6/config.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6/special_tokens_map.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6/tokenizer_config.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6/tokenizer.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama-3-8b-ok-5775-v6/flywheel_model.0.safetensors
hastagaras-llama-3-8b-ok-5775-v6-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-v6-mkmlizer: warnings.warn(
hastagaras-llama-3-8b-ok-5775-v6-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-v6-mkmlizer: warnings.warn(
hastagaras-llama-3-8b-ok-5775-v6-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-v6-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Saving duration: 0.247s
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.291s
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/tokenizer_config.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/merges.txt
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/special_tokens_map.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/config.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/vocab.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/tokenizer.json
hastagaras-llama-3-8b-ok-5775-v6-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama-3-8b-ok-5775-v6_reward/reward.tensors
Job hastagaras-llama-3-8b-ok-5775-v6-mkmlizer completed after 62.86s with status: succeeded
Stopping job with name hastagaras-llama-3-8b-ok-5775-v6-mkmlizer
Pipeline stage MKMLizer completed in 69.21s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama-3-8b-ok-5775-v6
Waiting for inference service hastagaras-llama-3-8b-ok-5775-v6 to be ready
Inference service hastagaras-llama-3-8b-ok-5775-v6 ready after 40.233362913131714s
Pipeline stage ISVCDeployer completed in 47.84s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1870293617248535s
Received healthy response to inference request in 1.3650553226470947s
Received healthy response to inference request in 1.3203399181365967s
Received healthy response to inference request in 1.3038654327392578s
Received healthy response to inference request in 1.556257963180542s
5 requests
0 failed requests
5th percentile: 1.3071603298187255
10th percentile: 1.3104552268981933
20th percentile: 1.317045021057129
30th percentile: 1.3292829990386963
40th percentile: 1.3471691608428955
50th percentile: 1.3650553226470947
60th percentile: 1.4415363788604736
70th percentile: 1.5180174350738525
80th percentile: 1.6824122428894044
90th percentile: 1.934720802307129
95th percentile: 2.060875082015991
99th percentile: 2.161798505783081
mean time: 1.546509599685669
Pipeline stage StressChecker completed in 9.01s
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-llama-3-8b-ok_5775_v6 status is now deployed due to DeploymentManager action
hastagaras-llama-3-8b-ok_5775_v6 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics