submission_id: nousresearch-meta-llama_4941_v53
developer_uid: zonemercy
status: inactive
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, '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': "Bot's Name: {bot_name}\nBot's Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', '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: 2024-05-04T18:06:25+00:00
model_name: nousresearch-meta-llama_4941_v53
model_eval_status: success
double_thumbs_up: 535
thumbs_up: 818
thumbs_down: 512
num_battles: 41747
num_wins: 21502
celo_rating: 1179.81
entertaining: 7.14
stay_in_character: 8.52
user_preference: 7.46
safety_score: 0.95
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: nousresearch-meta-llama_4941_v53
double_thumbs_up_ratio: 0.2868632707774799
feedback_count: 1865
ineligible_reason: None
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_score: 7.706666666666667
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.4386058981233244
thumbs_down_ratio: 0.27453083109919574
thumbs_up_ratio: 0.7254691689008043
us_pacific_date: 2024-05-04
win_ratio: 0.5150549740101085
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v53-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v53-mkmlizer to finish
nousresearch-meta-llama-4941-v53-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v53-mkmlizer: ║ _____ __ __ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ /___/ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ Version: 0.8.10 ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ The license key for the current software has been verified as ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ belonging to: ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ Chai Research Corp. ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
nousresearch-meta-llama-4941-v53-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v53-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v53-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.
nousresearch-meta-llama-4941-v53-mkmlizer: warnings.warn(warning_message, FutureWarning)
nousresearch-meta-llama-4941-v53-mkmlizer: Downloaded to shared memory in 13.943s
nousresearch-meta-llama-4941-v53-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v53-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v53-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 32.286s
nousresearch-meta-llama-4941-v53-mkmlizer: creating bucket guanaco-mkml-models
nousresearch-meta-llama-4941-v53-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v53-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53
nousresearch-meta-llama-4941-v53-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53/config.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53/tokenizer_config.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53/special_tokens_map.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53/tokenizer.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v53/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v53-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
nousresearch-meta-llama-4941-v53-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.
nousresearch-meta-llama-4941-v53-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v53-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.
nousresearch-meta-llama-4941-v53-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v53-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.
nousresearch-meta-llama-4941-v53-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v53-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()
nousresearch-meta-llama-4941-v53-mkmlizer: return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v53-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nousresearch-meta-llama-4941-v53-mkmlizer: Saving duration: 0.234s
nousresearch-meta-llama-4941-v53-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.749s
nousresearch-meta-llama-4941-v53-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v53-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v53-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/config.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/merges.txt
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/vocab.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/tokenizer.json
nousresearch-meta-llama-4941-v53-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v53_reward/reward.tensors
Job nousresearch-meta-llama-4941-v53-mkmlizer completed after 100.83s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v53-mkmlizer
Pipeline stage MKMLizer completed in 103.93s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v53
Waiting for inference service nousresearch-meta-llama-4941-v53 to be ready
Inference service nousresearch-meta-llama-4941-v53 ready after 50.33797907829285s
Pipeline stage ISVCDeployer completed in 62.68s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0807199478149414s
Received healthy response to inference request in 1.288067102432251s
Received healthy response to inference request in 1.251417875289917s
Received healthy response to inference request in 1.26957106590271s
Received healthy response to inference request in 1.268662691116333s
5 requests
0 failed requests
5th percentile: 1.2548668384552002
10th percentile: 1.2583158016204834
20th percentile: 1.2652137279510498
30th percentile: 1.2688443660736084
40th percentile: 1.2692077159881592
50th percentile: 1.26957106590271
60th percentile: 1.2769694805145264
70th percentile: 1.2843678951263429
80th percentile: 1.4465976715087892
90th percentile: 1.7636588096618653
95th percentile: 1.9221893787384032
99th percentile: 2.0490138339996338
mean time: 1.4316877365112304
Pipeline stage StressChecker completed in 7.88s
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 0.08s
nousresearch-meta-llama_4941_v53 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v53 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics