submission_id: r136a1-apocalypse_v1
developer_uid: R136a1
status: inactive
model_repo: R136a1/Apocalypse
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.1, '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': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{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: 2024-06-26T17:05:35+00:00
model_name: r136a1-apocalypse_v1
model_group: R136a1/Apocalypse
num_battles: 32607
num_wins: 17816
celo_rating: 1214.74
propriety_score: 0.7157646442143887
propriety_total_count: 15262.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: r136a1-apocalypse_v1
ineligible_reason: None
language_model: R136a1/Apocalypse
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-26
win_ratio: 0.5463857453920937
Resubmit model
Running pipeline stage MKMLizer
Starting job with name r136a1-apocalypse-v1-mkmlizer
Waiting for job on r136a1-apocalypse-v1-mkmlizer to finish
r136a1-apocalypse-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.
r136a1-apocalypse-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
Job r136a1-apocalypse-v1-mkmlizer completed after 31.11s with status: failed
Stopping job with name r136a1-apocalypse-v1-mkmlizer
%s, retrying in %s seconds...
Starting job with name r136a1-apocalypse-v1-mkmlizer
Waiting for job on r136a1-apocalypse-v1-mkmlizer to finish
r136a1-apocalypse-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
r136a1-apocalypse-v1-mkmlizer: ║ _____ __ __ ║
r136a1-apocalypse-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
r136a1-apocalypse-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
r136a1-apocalypse-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
r136a1-apocalypse-v1-mkmlizer: ║ /___/ ║
r136a1-apocalypse-v1-mkmlizer: ║ ║
r136a1-apocalypse-v1-mkmlizer: ║ Version: 0.8.14 ║
r136a1-apocalypse-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
r136a1-apocalypse-v1-mkmlizer: ║ https://mk1.ai ║
r136a1-apocalypse-v1-mkmlizer: ║ ║
r136a1-apocalypse-v1-mkmlizer: ║ The license key for the current software has been verified as ║
r136a1-apocalypse-v1-mkmlizer: ║ belonging to: ║
r136a1-apocalypse-v1-mkmlizer: ║ ║
r136a1-apocalypse-v1-mkmlizer: ║ Chai Research Corp. ║
r136a1-apocalypse-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
r136a1-apocalypse-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
r136a1-apocalypse-v1-mkmlizer: ║ ║
r136a1-apocalypse-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
r136a1-apocalypse-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.
r136a1-apocalypse-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
r136a1-apocalypse-v1-mkmlizer: Downloaded to shared memory in 47.518s
r136a1-apocalypse-v1-mkmlizer: quantizing model to /dev/shm/model_cache
r136a1-apocalypse-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
r136a1-apocalypse-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:00, 2.08s/it] Loading 0: 5%|▌ | 15/291 [00:04<00:57, 4.77it/s] Loading 0: 11%|█ | 32/291 [00:04<00:21, 12.22it/s] Loading 0: 17%|█▋ | 49/291 [00:04<00:11, 21.76it/s] Loading 0: 22%|██▏ | 63/291 [00:04<00:09, 24.48it/s] Loading 0: 27%|██▋ | 78/291 [00:05<00:06, 34.46it/s] Loading 0: 33%|███▎ | 95/291 [00:05<00:04, 48.49it/s] Loading 0: 38%|███▊ | 112/291 [00:05<00:02, 64.16it/s] Loading 0: 45%|████▍ | 130/291 [00:05<00:02, 80.47it/s] Loading 0: 51%|█████ | 148/291 [00:05<00:01, 96.01it/s] Loading 0: 56%|█████▋ | 164/291 [00:05<00:01, 108.70it/s] Loading 0: 62%|██████▏ | 180/291 [00:05<00:01, 68.47it/s] Loading 0: 67%|██████▋ | 195/291 [00:06<00:01, 79.16it/s] Loading 0: 73%|███████▎ | 212/291 [00:06<00:00, 94.37it/s] Loading 0: 79%|███████▊ | 229/291 [00:06<00:00, 108.61it/s] Loading 0: 84%|████████▍ | 244/291 [00:06<00:00, 116.90it/s] Loading 0: 89%|████████▉ | 259/291 [00:06<00:00, 122.20it/s] Loading 0: 94%|█████████▍| 274/291 [00:06<00:00, 73.53it/s] Loading 0: 99%|█████████▉| 289/291 [00:07<00:00, 86.05it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
r136a1-apocalypse-v1-mkmlizer: quantized model in 19.098s
r136a1-apocalypse-v1-mkmlizer: Processed model R136a1/Apocalypse in 67.701s
r136a1-apocalypse-v1-mkmlizer: creating bucket guanaco-mkml-models
r136a1-apocalypse-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/r136a1-apocalypse-v1/config.json
r136a1-apocalypse-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/r136a1-apocalypse-v1/special_tokens_map.json
r136a1-apocalypse-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/r136a1-apocalypse-v1/tokenizer_config.json
r136a1-apocalypse-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/r136a1-apocalypse-v1/tokenizer.json
r136a1-apocalypse-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/r136a1-apocalypse-v1/flywheel_model.0.safetensors
r136a1-apocalypse-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
r136a1-apocalypse-v1-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.
r136a1-apocalypse-v1-mkmlizer: warnings.warn(
r136a1-apocalypse-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.
r136a1-apocalypse-v1-mkmlizer: warnings.warn(
r136a1-apocalypse-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.
r136a1-apocalypse-v1-mkmlizer: warnings.warn(
r136a1-apocalypse-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()
r136a1-apocalypse-v1-mkmlizer: return self.fget.__get__(instance, owner)()
r136a1-apocalypse-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
r136a1-apocalypse-v1-mkmlizer: Saving duration: 0.313s
r136a1-apocalypse-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.837s
r136a1-apocalypse-v1-mkmlizer: creating bucket guanaco-reward-models
r136a1-apocalypse-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
r136a1-apocalypse-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/r136a1-apocalypse-v1_reward
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/config.json
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/special_tokens_map.json
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/tokenizer_config.json
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/merges.txt
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/vocab.json
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/tokenizer.json
r136a1-apocalypse-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/r136a1-apocalypse-v1_reward/reward.tensors
Job r136a1-apocalypse-v1-mkmlizer completed after 95.2s with status: succeeded
Stopping job with name r136a1-apocalypse-v1-mkmlizer
Pipeline stage MKMLizer completed in 127.39s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service r136a1-apocalypse-v1
Waiting for inference service r136a1-apocalypse-v1 to be ready
Inference service r136a1-apocalypse-v1 ready after 40.23862409591675s
Pipeline stage ISVCDeployer completed in 46.74s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1622066497802734s
Received healthy response to inference request in 1.335822343826294s
Received healthy response to inference request in 1.308488130569458s
Received healthy response to inference request in 1.2674808502197266s
Received healthy response to inference request in 1.351060152053833s
5 requests
0 failed requests
5th percentile: 1.275682306289673
10th percentile: 1.283883762359619
20th percentile: 1.3002866744995116
30th percentile: 1.313954973220825
40th percentile: 1.3248886585235595
50th percentile: 1.335822343826294
60th percentile: 1.3419174671173095
70th percentile: 1.348012590408325
80th percentile: 1.5132894515991213
90th percentile: 1.8377480506896973
95th percentile: 1.9999773502349851
99th percentile: 2.1297607898712156
mean time: 1.485011625289917
Pipeline stage StressChecker completed in 8.19s
r136a1-apocalypse_v1 status is now deployed due to DeploymentManager action
r136a1-apocalypse_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics