submission_id: thedrummer-llama-3some-8_2258_v1
developer_uid: Meliodia
status: rejected
model_repo: TheDrummer/Llama-3SOME-8B-v1-BETA
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_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}
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: 2024-04-27T20:39:39+00:00
model_name: thedrummer-llama-3some-8_2258_v1
model_eval_status: success
double_thumbs_up: 0
thumbs_up: 1
thumbs_down: 7
num_battles: 416
num_wins: 47
celo_rating: None
entertaining: 1.06
stay_in_character: 1.17
user_preference: 1.12
safety_score: None
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: thedrummer-llama-3some-8_2258_v1
double_thumbs_up_ratio: 0.0
feedback_count: 8
ineligible_reason: model is not deployable
language_model: TheDrummer/Llama-3SOME-8B-v1-BETA
model_score: 1.1166666666666667
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.125
thumbs_down_ratio: 0.875
thumbs_up_ratio: 0.125
us_pacific_date: 2024-04-27
win_ratio: 0.11298076923076923
Resubmit model
Running pipeline stage MKMLizer
Starting job with name thedrummer-llama-3some-8-2258-v1-mkmlizer
Waiting for job on thedrummer-llama-3some-8-2258-v1-mkmlizer to finish
Stopping job with name thedrummer-llama-3some-8-2258-v1-mkmlizer
%s, retrying in %s seconds...
Starting job with name thedrummer-llama-3some-8-2258-v1-mkmlizer
Waiting for job on thedrummer-llama-3some-8-2258-v1-mkmlizer to finish
thedrummer-llama-3some-8-2258-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ _____ __ __ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ /___/ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ Version: 0.8.10 ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ The license key for the current software has been verified as ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ belonging to: ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ Chai Research Corp. ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ║ ║
thedrummer-llama-3some-8-2258-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
thedrummer-llama-3some-8-2258-v1-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.
thedrummer-llama-3some-8-2258-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
thedrummer-llama-3some-8-2258-v1-mkmlizer: Downloaded to shared memory in 30.007s
thedrummer-llama-3some-8-2258-v1-mkmlizer: quantizing model to /dev/shm/model_cache
thedrummer-llama-3some-8-2258-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
thedrummer-llama-3some-8-2258-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:02, 82.21it/s] Loading 0: 99%|█████████▊| 287/291 [00:02<00:00, 129.62it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
thedrummer-llama-3some-8-2258-v1-mkmlizer: quantized model in 18.858s
thedrummer-llama-3some-8-2258-v1-mkmlizer: Processed model TheDrummer/Llama-3SOME-8B-v1-BETA in 49.894s
thedrummer-llama-3some-8-2258-v1-mkmlizer: creating bucket guanaco-mkml-models
thedrummer-llama-3some-8-2258-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
thedrummer-llama-3some-8-2258-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1/config.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1/special_tokens_map.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1/tokenizer_config.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1/tokenizer.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/thedrummer-llama-3some-8-2258-v1/flywheel_model.0.safetensors
thedrummer-llama-3some-8-2258-v1-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.
thedrummer-llama-3some-8-2258-v1-mkmlizer: warnings.warn(
thedrummer-llama-3some-8-2258-v1-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.
thedrummer-llama-3some-8-2258-v1-mkmlizer: warnings.warn(
thedrummer-llama-3some-8-2258-v1-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()
thedrummer-llama-3some-8-2258-v1-mkmlizer: return self.fget.__get__(instance, owner)()
thedrummer-llama-3some-8-2258-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
thedrummer-llama-3some-8-2258-v1-mkmlizer: Saving duration: 0.296s
thedrummer-llama-3some-8-2258-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.444s
thedrummer-llama-3some-8-2258-v1-mkmlizer: creating bucket guanaco-reward-models
thedrummer-llama-3some-8-2258-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
thedrummer-llama-3some-8-2258-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/config.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/special_tokens_map.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/tokenizer_config.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/merges.txt
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/vocab.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/tokenizer.json
thedrummer-llama-3some-8-2258-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/thedrummer-llama-3some-8-2258-v1_reward/reward.tensors
Job thedrummer-llama-3some-8-2258-v1-mkmlizer completed after 86.38s with status: succeeded
Stopping job with name thedrummer-llama-3some-8-2258-v1-mkmlizer
Pipeline stage MKMLizer completed in 94.48s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service thedrummer-llama-3some-8-2258-v1
Waiting for inference service thedrummer-llama-3some-8-2258-v1 to be ready
Inference service thedrummer-llama-3some-8-2258-v1 ready after 30.32249617576599s
Pipeline stage ISVCDeployer completed in 38.01s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0839014053344727s
Received healthy response to inference request in 1.220618486404419s
Received healthy response to inference request in 1.256225824356079s
Received healthy response to inference request in 1.2807397842407227s
Received healthy response to inference request in 1.2646605968475342s
5 requests
0 failed requests
5th percentile: 1.2277399539947509
10th percentile: 1.234861421585083
20th percentile: 1.2491043567657472
30th percentile: 1.25791277885437
40th percentile: 1.261286687850952
50th percentile: 1.2646605968475342
60th percentile: 1.2710922718048097
70th percentile: 1.277523946762085
80th percentile: 1.4413721084594728
90th percentile: 1.7626367568969727
95th percentile: 1.9232690811157225
99th percentile: 2.0517749404907226
mean time: 1.4212292194366456
Pipeline stage StressChecker completed in 7.75s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
%s, retrying in %s seconds...
thedrummer-llama-3some-8_2258_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
%s, retrying in %s seconds...
thedrummer-llama-3some-8_2258_v1 status is now rejected due to its M-Eval score being less than the acceptable minimum 6.5 to serve to users. Please consider iterating on your model's ability to adhere to prompts to improve this score.

Usage Metrics

Latency Metrics