submission_id: orenguteng-llama-3-8b-le_8683_v2
developer_uid: robert_irvine
status: inactive
model_repo: Orenguteng/Llama-3-8B-Lexi-Uncensored
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>', '<|start_header_id|>', '\n\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': '<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nThis is an entertaining conversation. You are {bot_name} who has persona: {memory}. Engage in a chat with {user_name} while staying in character. Try to flirt with {user_name}. Engage in *roleplay* actions. Describe the scene dramatically<|eot_id|>', 'prompt_template': '<|start_header_id|>system<|end_header_id|>\n\nExample conversation:\n{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': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-05-02T16:54:41+00:00
model_name: orenguteng-llama-3-8b-le_8683_v2
model_eval_status: pending
double_thumbs_up: 57
thumbs_up: 88
thumbs_down: 68
num_battles: 7658
num_wins: 3531
celo_rating: 1136.14
entertaining: None
stay_in_character: None
user_preference: None
safety_score: None
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: orenguteng-llama-3-8b-le_8683_v2
double_thumbs_up_ratio: 0.2676056338028169
feedback_count: 213
ineligible_reason: None
language_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
model_score: None
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.4131455399061033
thumbs_down_ratio: 0.3192488262910798
thumbs_up_ratio: 0.6807511737089202
us_pacific_date: 2024-05-02
win_ratio: 0.46108644554714023
Resubmit model
Running pipeline stage MKMLizer
Starting job with name orenguteng-llama-3-8b-le-8683-v2-mkmlizer
Waiting for job on orenguteng-llama-3-8b-le-8683-v2-mkmlizer to finish
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ _____ __ __ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ /___/ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ Version: 0.8.10 ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ The license key for the current software has been verified as ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ belonging to: ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ Chai Research Corp. ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ║ ║
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
orenguteng-llama-3-8b-le-8683-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.
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Downloaded to shared memory in 26.194s
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: quantizing model to /dev/shm/model_cache
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 48%|████▊ | 139/291 [00:01<00:01, 138.92it/s] Loading 0: 89%|████████▉ | 259/291 [00:06<00:00, 34.04it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: quantized model in 17.222s
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Processed model Orenguteng/Llama-3-8B-Lexi-Uncensored in 44.420s
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: creating bucket guanaco-mkml-models
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2/config.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2/special_tokens_map.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2/tokenizer_config.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2/tokenizer.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/orenguteng-llama-3-8b-le-8683-v2/flywheel_model.0.safetensors
orenguteng-llama-3-8b-le-8683-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.
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: warnings.warn(
orenguteng-llama-3-8b-le-8683-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()
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: return self.fget.__get__(instance, owner)()
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Saving duration: 0.247s
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 5.638s
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: creating bucket guanaco-reward-models
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/config.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/special_tokens_map.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/tokenizer_config.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/vocab.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/merges.txt
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/tokenizer.json
orenguteng-llama-3-8b-le-8683-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/orenguteng-llama-3-8b-le-8683-v2_reward/reward.tensors
Job orenguteng-llama-3-8b-le-8683-v2-mkmlizer completed after 67.65s with status: succeeded
Stopping job with name orenguteng-llama-3-8b-le-8683-v2-mkmlizer
Pipeline stage MKMLizer completed in 68.89s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.29s
Running pipeline stage ISVCDeployer
Creating inference service orenguteng-llama-3-8b-le-8683-v2
Waiting for inference service orenguteng-llama-3-8b-le-8683-v2 to be ready
Inference service orenguteng-llama-3-8b-le-8683-v2 ready after 40.61459279060364s
Pipeline stage ISVCDeployer completed in 46.80s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.152100086212158s
Received healthy response to inference request in 1.192354679107666s
Received healthy response to inference request in 1.2012557983398438s
Received healthy response to inference request in 1.189082384109497s
Received healthy response to inference request in 1.1988797187805176s
5 requests
0 failed requests
5th percentile: 1.1897368431091309
10th percentile: 1.1903913021087646
20th percentile: 1.1917002201080322
30th percentile: 1.1936596870422362
40th percentile: 1.196269702911377
50th percentile: 1.1988797187805176
60th percentile: 1.1998301506042481
70th percentile: 1.2007805824279785
80th percentile: 1.3914246559143069
90th percentile: 1.7717623710632324
95th percentile: 1.961931228637695
99th percentile: 2.114066314697266
mean time: 1.3867345333099366
Pipeline stage StressChecker completed in 9.34s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.21s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.12s
Running M-Eval for topic stay_in_character
%s, retrying in %s seconds...
orenguteng-llama-3-8b-le_8683_v2 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
%s, retrying in %s seconds...
orenguteng-llama-3-8b-le_8683_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics