submission_id: thetsar1209-llama3-carp-v0-4_v2
developer_uid: TheTsar1209
status: inactive
model_repo: TheTsar1209/llama3-carp-v0.4
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['<|im_end|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': '<|im_start|>system\nYou are {bot_name}\nDescription : {memory}\n', 'prompt_template': 'Example conversation:\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', 'truncate_by_message': False}
reward_formatter: {'memory_template': '<|im_start|>system\nYou are {bot_name}\nDescription : {memory}\n', 'prompt_template': 'Example conversation:\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', 'truncate_by_message': False}
timestamp: 2024-05-09T00:18:01+00:00
model_name: thetsar1209-llama3-carp-v0-4_v1
model_eval_status: success
double_thumbs_up: 304
thumbs_up: 426
thumbs_down: 203
num_battles: 20421
num_wins: 10389
celo_rating: 1182.46
entertaining: 7.16
stay_in_character: 8.58
user_preference: 7.3
safety_score: 0.96
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: thetsar1209-llama3-carp-v0-4_v1
double_thumbs_up_ratio: 0.3258306538049303
feedback_count: 933
ineligible_reason: None
language_model: TheTsar1209/llama3-carp-v0.4
model_score: 7.68
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.4565916398713826
thumbs_down_ratio: 0.21757770632368703
thumbs_up_ratio: 0.782422293676313
us_pacific_date: 2024-05-08
win_ratio: 0.5087410019097988
Resubmit model
Running pipeline stage MKMLizer
Starting job with name thetsar1209-llama3-carp-v0-4-v2-mkmlizer
Waiting for job on thetsar1209-llama3-carp-v0-4-v2-mkmlizer to finish
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ _____ __ __ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ /___/ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ Version: 0.8.10 ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ The license key for the current software has been verified as ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ belonging to: ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ Chai Research Corp. ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ║ ║
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
thetsar1209-llama3-carp-v0-4-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.
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Downloaded to shared memory in 14.310s
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: quantizing model to /dev/shm/model_cache
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:02, 74.58it/s] Loading 0: 64%|██████▍ | 187/291 [00:02<00:01, 79.67it/s] Loading 0: 98%|█████████▊| 284/291 [00:03<00:00, 86.65it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: quantized model in 20.729s
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Processed model TheTsar1209/llama3-carp-v0.4 in 36.126s
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: creating bucket guanaco-mkml-models
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2/config.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2/special_tokens_map.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2/tokenizer_config.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2/tokenizer.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/thetsar1209-llama3-carp-v0-4-v2/flywheel_model.0.safetensors
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
thetsar1209-llama3-carp-v0-4-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.
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: warnings.warn(
thetsar1209-llama3-carp-v0-4-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.
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: warnings.warn(
thetsar1209-llama3-carp-v0-4-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.
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: warnings.warn(
thetsar1209-llama3-carp-v0-4-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()
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: return self.fget.__get__(instance, owner)()
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Saving duration: 0.264s
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.295s
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: creating bucket guanaco-reward-models
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/special_tokens_map.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/config.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/tokenizer_config.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/merges.txt
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/vocab.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/tokenizer.json
thetsar1209-llama3-carp-v0-4-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/thetsar1209-llama3-carp-v0-4-v2_reward/reward.tensors
Job thetsar1209-llama3-carp-v0-4-v2-mkmlizer completed after 63.12s with status: succeeded
Stopping job with name thetsar1209-llama3-carp-v0-4-v2-mkmlizer
Pipeline stage MKMLizer completed in 66.10s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service thetsar1209-llama3-carp-v0-4-v2
Waiting for inference service thetsar1209-llama3-carp-v0-4-v2 to be ready
Inference service thetsar1209-llama3-carp-v0-4-v2 ready after 30.189130067825317s
Pipeline stage ISVCDeployer completed in 37.20s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1624889373779297s
Received healthy response to inference request in 1.2970342636108398s
Received healthy response to inference request in 1.2855632305145264s
Received healthy response to inference request in 1.2948596477508545s
Received healthy response to inference request in 1.291841745376587s
5 requests
0 failed requests
5th percentile: 1.2868189334869384
10th percentile: 1.2880746364593505
20th percentile: 1.2905860424041748
30th percentile: 1.2924453258514403
40th percentile: 1.2936524868011474
50th percentile: 1.2948596477508545
60th percentile: 1.2957294940948487
70th percentile: 1.2965993404388427
80th percentile: 1.470125198364258
90th percentile: 1.8163070678710938
95th percentile: 1.9893980026245115
99th percentile: 2.1278707504272463
mean time: 1.4663575649261475
Pipeline stage StressChecker completed in 7.93s
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 0.03s
M-Eval Dataset for topic stay_in_character is loaded
thetsar1209-llama3-carp-v0-4_v2 status is now deployed due to DeploymentManager action
thetsar1209-llama3-carp-v0-4_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics