jellywibble-lora-120k-pr_2827

developer_uid: robert_irvine

submission_id: jellywibble-lora-120k-pr_2827_v8

model_name: jellywibble-lora-120k-pr_2827_v8

model_group: Jellywibble/lora_120k_pr

status: torndown

timestamp: 2024-07-06T06:08:07+00:00

num_battles: 36439

num_wins: 17552

celo_rating: 1170.76

family_friendly_score: 0.0

submission_type: basic

model_repo: Jellywibble/lora_120k_pref_data_ep2

model_architecture: LlamaForCausalLM

reward_repo: Jellywibble/CHAI_alignment_reward_model

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: jellywibble-lora-120k-pr_2827_v8

is_internal_developer: True

language_model: Jellywibble/lora_120k_pref_data_ep2

model_size: 8B

ranking_group: single

us_pacific_date: 2024-07-05

win_ratio: 0.48168171464639536

generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], '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: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}

Resubmit model

Running pipeline stage MKMLizer
Starting job with name jellywibble-lora-120k-pr-2827-v8-mkmlizer
Waiting for job on jellywibble-lora-120k-pr-2827-v8-mkmlizer to finish
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║     _____            __           __                                ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║       /___/                                                         ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  Version: 0.8.14                                                    ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  https://mk1.ai                                                     ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  The license key for the current software has been verified as      ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  belonging to:                                                      ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  Chai Research Corp.                                                ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v8-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Downloaded to shared memory in 40.993s
jellywibble-lora-120k-pr-2827-v8-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jellywibble-lora-120k-pr-2827-v8-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-lora-120k-pr-2827-v8-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8/tokenizer_config.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8/special_tokens_map.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8/config.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8/tokenizer.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v8/flywheel_model.0.safetensors
jellywibble-lora-120k-pr-2827-v8-mkmlizer: loading reward model from Jellywibble/CHAI_alignment_reward_model
jellywibble-lora-120k-pr-2827-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-lora-120k-pr-2827-v8-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
jellywibble-lora-120k-pr-2827-v8-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-lora-120k-pr-2827-v8-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v8-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.
jellywibble-lora-120k-pr-2827-v8-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Saving duration: 0.147s
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Processed model Jellywibble/CHAI_alignment_reward_model in 2.428s
jellywibble-lora-120k-pr-2827-v8-mkmlizer: creating bucket guanaco-reward-models
jellywibble-lora-120k-pr-2827-v8-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jellywibble-lora-120k-pr-2827-v8-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/config.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/special_tokens_map.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/tokenizer_config.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/merges.txt
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/vocab.json
jellywibble-lora-120k-pr-2827-v8-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v8_reward/tokenizer.json
Job jellywibble-lora-120k-pr-2827-v8-mkmlizer completed after 109.28s with status: succeeded
Stopping job with name jellywibble-lora-120k-pr-2827-v8-mkmlizer
Pipeline stage MKMLizer completed in 110.13s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-lora-120k-pr-2827-v8
Waiting for inference service jellywibble-lora-120k-pr-2827-v8 to be ready
Failed to get response for submission azazelle-l3-persephone-8_4975_v3: ('http://azazelle-l3-persephone-8-4975-v3-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:47766->127.0.0.1:8080: read: connection reset by peer\n')
Inference service jellywibble-lora-120k-pr-2827-v8 ready after 50.23133969306946s
Pipeline stage ISVCDeployer completed in 57.31s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.072497606277466s
Received healthy response to inference request in 1.3438401222229004s
Received healthy response to inference request in 1.3292686939239502s
Received healthy response to inference request in 1.361119031906128s
Received healthy response to inference request in 1.3433845043182373s
5 requests
0 failed requests
5th percentile: 1.3320918560028077
10th percentile: 1.3349150180816651
20th percentile: 1.3405613422393798
30th percentile: 1.34347562789917
40th percentile: 1.343657875061035
50th percentile: 1.3438401222229004
60th percentile: 1.3507516860961915
70th percentile: 1.3576632499694825
80th percentile: 1.5033947467803956
90th percentile: 1.7879461765289308
95th percentile: 1.930221891403198
99th percentile: 2.0440424633026124
mean time: 1.4900219917297364
Pipeline stage StressChecker completed in 8.07s
jellywibble-lora-120k-pr_2827_v8 status is now deployed due to DeploymentManager action
jellywibble-lora-120k-pr_2827_v8 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jellywibble-lora-120k-pr_2827_v8
Running pipeline stage ISVCDeleter
Checking if service jellywibble-lora-120k-pr-2827-v8 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.35s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2827-v8/config.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v8/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v8/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v8/tokenizer.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v8/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/merges.txt from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v8_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.28s
jellywibble-lora-120k-pr_2827_v8 status is now torndown due to DeploymentManager action