dphn/dolphin-2.9.4-llama3.1-8b-gguf

Name: dphn/dolphin-2.9.4-llama3.1-8b-gguf
Author: dphn

High-quality GGUF model

3.1K 📥 Downloads

73 ❤️ Likes

12 📁 GGUF Files

57.3 GB 💾 Total Size

2 years ago 🔄 Last Updated

📋 Model Description

license: llama3.1 base_model: meta-llama/Meta-Llama-3.1-8B tags:

generatedfromtrainer

datasets:

cognitivecomputations/Dolphin-2.9
m-a-p/CodeFeedback-Filtered-Instruction
cognitivecomputations/dolphin-coder
cognitivecomputations/samantha-data
microsoft/orca-math-word-problems-200k
mlabonne/FineTome-100k
arcee/agent_data
PawanKrd/math-gpt-4o-200k
cognitivecomputations/SystemChat-2.0

Dolphin 2.9.4 Llama 3.1 8b 🐬

This is the GGUF conversion, for use with llama.cpp, ollama, lmstudio etc.

Curated and trained by Eric Hartford and Cognitive Computations

Discord</a>
Discord: https://discord.gg/h3K4XGj2RH

Our appreciation for the sponsors of Dolphin 2.9.4:

Crusoe Cloud - provided excellent on-demand 8xL40S node

This model is based on Meta Llama 3.1 8b, and is governed by the Llama 3.1 license.

The base model has 128K context, and our finetuning used 8192 sequence length.

Dolphin 2.9.4 uses ChatML prompt template format.

example:

<|im_start|>system
You are Dolphin, a helpful AI assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Dolphin-2.9.4 has a variety of instruction following, conversational, and coding skills. It also has agentic abilities and supports function calling.
It is especially trained to obey the system prompt, and follow instructions in many languages.

Dolphin is uncensored. We have filtered the dataset to remove alignment and bias. This makes the model more compliant. You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant with any requests, even unethical ones. Please read my blog post about uncensored models. https://erichartford.com/uncensored-models You are responsible for any content you create using this model. Enjoy responsibly.

Evals

hf (pretrained=/workspace/axolotl/dolphin-2.9.4-llama3.1-8b-hf,dtype=bfloat16), genkwargs: (None), limit: None, numfewshot: None, batch_size: auto (4)

                Tasks Version Filter n-shot Metric Value Stderr
                leaderboard N/A none 0 acc ↑ 0.2926 ± 0.0041
none 0 acc_norm ↑ 0.4513 ± 0.0053
none 0 exact_match ↑ 0.0982 ± 0.0079
none 0 instlevelloose_acc ↑ 0.3825 ± N/A
none 0 instlevelstrict_acc ↑ 0.3597 ± N/A
none 0 promptlevelloose_acc ↑ 0.2421 ± 0.0184
none 0 promptlevelstrict_acc ↑ 0.2181 ± 0.0178
- leaderboardbbh N/A none 3 accnorm ↑ 0.4931 ± 0.0061
- leaderboardbbhbooleanexpressions 0 none 3 accnorm ↑ 0.8000 ± 0.0253
- leaderboardbbhcausaljudgement 0 none 3 accnorm ↑ 0.5615 ± 0.0364
- leaderboardbbhdateunderstanding 0 none 3 accnorm ↑ 0.4520 ± 0.0315
- leaderboardbbhdisambiguationqa 0 none 3 accnorm ↑ 0.6640 ± 0.0299
- leaderboardbbhformalfallacies 0 none 3 accnorm ↑ 0.5600 ± 0.0315
- leaderboardbbhgeometricshapes 0 none 3 accnorm ↑ 0.3640 ± 0.0305
- leaderboardbbhhyperbaton 0 none 3 acc_norm ↑ 0.6320 ± 0.0306
- leaderboardbbhlogicaldeductionfiveobjects 0 none 3 accnorm ↑ 0.4600 ± 0.0316
- leaderboardbbhlogicaldeductionsevenobjects 0 none 3 accnorm ↑ 0.4360 ± 0.0314
- leaderboardbbhlogicaldeductionthreeobjects 0 none 3 accnorm ↑ 0.6160 ± 0.0308
- leaderboardbbhmovierecommendation 0 none 3 accnorm ↑ 0.7880 ± 0.0259
- leaderboardbbhnavigate 0 none 3 acc_norm ↑ 0.5200 ± 0.0317
- leaderboardbbhobjectcounting 0 none 3 accnorm ↑ 0.4520 ± 0.0315
- leaderboardbbhpenguinsinatable 0 none 3 accnorm ↑ 0.5205 ± 0.0415
- leaderboardbbhreasoningaboutcoloredobjects 0 none 3 accnorm ↑ 0.5120 ± 0.0317
- leaderboardbbhruinnames 0 none 3 accnorm ↑ 0.6320 ± 0.0306
- leaderboardbbhsalienttranslationerrordetection 0 none 3 accnorm ↑ 0.4320 ± 0.0314
- leaderboardbbhsnarks 0 none 3 acc_norm ↑ 0.5843 ± 0.0370
- leaderboardbbhsportsunderstanding 0 none 3 accnorm ↑ 0.7040 ± 0.0289
- leaderboardbbhtemporalsequences 0 none 3 accnorm ↑ 0.1440 ± 0.0222
- leaderboardbbhtrackingshuffledobjectsfiveobjects 0 none 3 acc_norm ↑ 0.1560 ± 0.0230
- leaderboardbbhtrackingshuffledobjectssevenobjects 0 none 3 acc_norm ↑ 0.1320 ± 0.0215
- leaderboardbbhtrackingshuffledobjectsthreeobjects 0 none 3 acc_norm ↑ 0.2840 ± 0.0286
- leaderboardbbhweboflies 0 none 3 acc_norm ↑ 0.4840 ± 0.0317
- leaderboardgpqa N/A none 0 accnorm ↑ 0.2903 ± 0.0132
- leaderboardgpqadiamond 1 none 0 acc_norm ↑ 0.2980 ± 0.0326
- leaderboardgpqaextended 1 none 0 acc_norm ↑ 0.2839 ± 0.0193
- leaderboardgpqamain 1 none 0 acc_norm ↑ 0.2946 ± 0.0216
- leaderboardifeval 2 none 0 instlevellooseacc ↑ 0.3825 ± N/A
none 0 instlevelstrict_acc ↑ 0.3597 ± N/A
none 0 promptlevelloose_acc ↑ 0.2421 ± 0.0184
none 0 promptlevelstrict_acc ↑ 0.2181 ± 0.0178
- leaderboardmathalgebrahard 1 none 4 exactmatch ↑ 0.1596 ± 0.0209
- leaderboardmathcountingandprobhard 1 none 4 exactmatch ↑ 0.0488 ± 0.0195
- leaderboardmathgeometryhard 1 none 4 exactmatch ↑ 0.0530 ± 0.0196
- leaderboardmathhard N/A none 4 exact_match ↑ 0.0982 ± 0.0079
- leaderboardmathintermediatealgebrahard 1 none 4 exact_match ↑ 0.0143 ± 0.0071
- leaderboardmathnumtheoryhard 1 none 4 exact_match ↑ 0.0455 ± 0.0168
- leaderboardmathprealgebrahard 1 none 4 exactmatch ↑ 0.2591 ± 0.0316
- leaderboardmathprecalculushard 1 none 4 exactmatch ↑ 0.0519 ± 0.0192
- leaderboardmmlupro 0.1 none 5 acc ↑ 0.2926 ± 0.0041
- leaderboardmusr N/A none 0 accnorm ↑ 0.3862 ± 0.0173
- leaderboardmusrmurdermysteries 1 none 0 accnorm ↑ 0.5280 ± 0.0316
- leaderboardmusrobjectplacements 1 none 0 accnorm ↑ 0.3594 ± 0.0300
- leaderboardmusrteamallocation 1 none 0 accnorm ↑ 0.2720 ± 0.0282
            

                Groups Version Filter n-shot Metric Value Stderr
                leaderboard N/A none 0 acc ↑ 0.2926 ± 0.0041
none 0 acc_norm ↑ 0.4513 ± 0.0053
none 0 exact_match ↑ 0.0982 ± 0.0079
none 0 instlevelloose_acc ↑ 0.3825 ± N/A
none 0 instlevelstrict_acc ↑ 0.3597 ± N/A
none 0 promptlevelloose_acc ↑ 0.2421 ± 0.0184
none 0 promptlevelstrict_acc ↑ 0.2181 ± 0.0178
- leaderboardbbh N/A none 3 accnorm ↑ 0.4931 ± 0.0061
- leaderboardgpqa N/A none 0 accnorm ↑ 0.2903 ± 0.0132
- leaderboardmathhard N/A none 4 exact_match ↑ 0.0982 ± 0.0079
- leaderboardmusr N/A none 0 accnorm ↑ 0.3862 ± 0.0173

See axolotl config

axolotl version: 0.4.1

base_model: meta-llama/Meta-Llama-3.1-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

loadin8bit: false
loadin4bit: true

strict: false

datasets:
  - path: /workspace/datasets/dolphin-2.9.4/dolphin201-sharegpt2.jsonl
    type: sharegpt
    conversation: chatml

chat_template: chatml
adapter: qlora

lora_r: 128

lora_alpha: 16

loramodulestosave: [embedtokens, lm_head]

lora_dropout: 0.05

loratargetlinear: true


unfrozen_parameters:
inputlayernorm
model.norm
postattentionlayernorm
selfattn.rotaryemb
^lmhead.weight$
^model.embed_tokens.weight$

mlp.down_proj layers

model.layers.1.mlp.downproj
model.layers.0.mlp.downproj
model.layers.30.mlp.downproj
model.layers.2.mlp.downproj
model.layers.21.mlp.downproj
model.layers.22.mlp.downproj
model.layers.29.mlp.downproj
model.layers.5.mlp.downproj
model.layers.4.mlp.downproj
model.layers.20.mlp.downproj
model.layers.23.mlp.downproj
model.layers.19.mlp.downproj
model.layers.3.mlp.downproj
model.layers.17.mlp.downproj
model.layers.6.mlp.downproj
model.layers.31.mlp.downproj

mlp.up_proj layers

model.layers.4.mlp.upproj
model.layers.3.mlp.upproj
model.layers.0.mlp.upproj
model.layers.5.mlp.upproj
model.layers.7.mlp.upproj
model.layers.6.mlp.upproj
model.layers.2.mlp.upproj
model.layers.1.mlp.upproj
model.layers.8.mlp.upproj
model.layers.12.mlp.upproj
model.layers.14.mlp.upproj
model.layers.9.mlp.upproj
model.layers.15.mlp.upproj
model.layers.17.mlp.upproj
model.layers.13.mlp.upproj
model.layers.19.mlp.upproj

selfattn.kproj layers

model.layers.29.selfattn.kproj
model.layers.25.selfattn.kproj
model.layers.23.selfattn.kproj
model.layers.28.selfattn.kproj
model.layers.21.selfattn.kproj
model.layers.19.selfattn.kproj
model.layers.22.selfattn.kproj
model.layers.20.selfattn.kproj
model.layers.24.selfattn.kproj
model.layers.31.selfattn.kproj
model.layers.27.selfattn.kproj
model.layers.26.selfattn.kproj
model.layers.17.selfattn.kproj
model.layers.11.selfattn.kproj
model.layers.18.selfattn.kproj
model.layers.14.selfattn.kproj

selfattn.oproj layers

model.layers.14.selfattn.oproj
model.layers.7.selfattn.oproj
model.layers.5.selfattn.oproj
model.layers.11.selfattn.oproj
model.layers.6.selfattn.oproj
model.layers.24.selfattn.oproj
model.layers.9.selfattn.oproj
model.layers.13.selfattn.oproj
model.layers.10.selfattn.oproj
model.layers.12.selfattn.oproj
model.layers.8.selfattn.oproj
model.layers.25.selfattn.oproj
model.layers.21.selfattn.oproj
model.layers.23.selfattn.oproj
model.layers.15.selfattn.oproj
model.layers.16.selfattn.oproj

selfattn.qproj layers

model.layers.8.selfattn.qproj
model.layers.13.selfattn.qproj
model.layers.9.selfattn.qproj
model.layers.14.selfattn.qproj
model.layers.10.selfattn.qproj
model.layers.11.selfattn.qproj
model.layers.0.selfattn.qproj
model.layers.15.selfattn.qproj
model.layers.1.selfattn.qproj
model.layers.6.selfattn.qproj
model.layers.5.selfattn.qproj
model.layers.7.selfattn.qproj
model.layers.12.selfattn.qproj
model.layers.16.selfattn.qproj
model.layers.17.selfattn.qproj
model.layers.26.selfattn.qproj

selfattn.vproj layers

model.layers.26.selfattn.vproj
model.layers.17.selfattn.vproj
model.layers.3.selfattn.vproj
model.layers.28.selfattn.vproj
model.layers.29.selfattn.vproj
model.layers.21.selfattn.vproj
model.layers.15.selfattn.vproj
model.layers.16.selfattn.vproj
model.layers.20.selfattn.vproj
model.layers.25.selfattn.vproj
model.layers.6.selfattn.vproj
model.layers.23.selfattn.vproj
model.layers.4.selfattn.vproj
model.layers.1.selfattn.vproj
model.layers.22.selfattn.vproj
model.layers.14.selfattn.vproj

mlp.gate_proj layers

model.layers.1.mlp.gateproj
model.layers.2.mlp.gateproj
model.layers.3.mlp.gateproj
model.layers.4.mlp.gateproj
model.layers.0.mlp.gateproj
model.layers.25.mlp.gateproj
model.layers.26.mlp.gateproj
model.layers.5.mlp.gateproj
model.layers.24.mlp.gateproj
model.layers.28.mlp.gateproj
model.layers.23.mlp.gateproj
model.layers.27.mlp.gateproj
model.layers.21.mlp.gateproj
model.layers.22.mlp.gateproj
model.layers.29.mlp.gateproj
model.layers.20.mlp.gateproj


datasetpreparedpath:  /workspace/axolotl/dolph-2.9.4-nemo-prepared
valsetsize: 0.01
output_dir: /workspace/axolotl/dolphin-2.9.4-llama3.1-8b

sequence_len: 8192
sample_packing: true
padtosequence_len: true

wandb_project: dolphin-2.9.4-llama3.1-8b
wandb_watch:
wandbrunid:
wandblogmodel:

gradientaccumulationsteps: 16
microbatchsize: 2
num_epochs: 3
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 5e-6
trainoninputs: false
groupbylength: false
bf16: auto
fp16:
tf32:

gradient_checkpointing: true
gradientcheckpointingkwargs:
  use_reentrant: false
earlystoppingpatience:
resumefromcheckpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 100
evalsperepoch: 4

evaltablesize:
savesperepoch: 1
savetotallimit: 2
save_steps:
debug:
deepspeed: deepspeedconfigs/zero3bf16.json
weight_decay: 0.1
special_tokens:
  eostoken: "<|imend|>"
  bostoken: "<|beginof_text|>"
  padtoken: "<|finetunerightpadid|>"
tokens:
  - "<|im_start|>"

fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdplimitall_gathers: true
  fsdpsyncmodule_states: true
  fsdpoffloadparams: true
  fsdpuseorig_params: false
  fsdpcpuramefficientloading: true
  fsdptransformerlayerclsto_wrap: MixtralSparseMoeBlock
  fsdpstatedicttype: FULLSTATE_DICT
  fsdpautowrappolicy: TRANSFORMERBASED_WRAP
  fsdpshardingstrategy: FULL_SHARD
  fsdpforwardprefetch: false
  fsdpbackwardprefetch: BACKWARD_PRE

workspace/axolotl/dolphin-2.9.4-llama3.1-8b

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the None dataset.
It achieves the following results on the evaluation set:

Loss: 0.5655

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learningrate: 5e-06
trainbatchsize: 2
evalbatchsize: 2
seed: 42
distributedtype: multi-GPU
numdevices: 8
gradientaccumulationsteps: 16
totaltrainbatchsize: 256
totalevalbatchsize: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lrschedulertype: cosine
lrschedulerwarmupsteps: 100
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.5837	1.0180	1161	0.5814
0.5525	2.0179	2322	0.5671
0.5514	2.9624	3420	0.5655

Framework versions

Transformers 4.44.0.dev0
Pytorch 2.4.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1

📂 GGUF File List

📁 Filename	📦 Size	⚡ Download
dolphin-2.9.4-llama3.1-8b-Q2_K.gguf LFS Q2	2.96 GB	Download
dolphin-2.9.4-llama3.1-8b-Q3_K_L.gguf LFS Q3	4.03 GB	Download
dolphin-2.9.4-llama3.1-8b-Q3_K_M.gguf LFS Q3	3.74 GB	Download
dolphin-2.9.4-llama3.1-8b-Q3_K_S.gguf LFS Q3	3.41 GB	Download
dolphin-2.9.4-llama3.1-8b-Q4_0.gguf Recommended LFS Q4	4.34 GB	Download
dolphin-2.9.4-llama3.1-8b-Q4_K_M.gguf LFS Q4	4.58 GB	Download
dolphin-2.9.4-llama3.1-8b-Q4_K_S.gguf LFS Q4	4.37 GB	Download
dolphin-2.9.4-llama3.1-8b-Q5_0.gguf LFS Q5	5.21 GB	Download
dolphin-2.9.4-llama3.1-8b-Q5_K_M.gguf LFS Q5	5.34 GB	Download
dolphin-2.9.4-llama3.1-8b-Q5_K_S.gguf LFS Q5	5.21 GB	Download
dolphin-2.9.4-llama3.1-8b-Q6_K.gguf LFS Q6	6.14 GB	Download
dolphin-2.9.4-llama3.1-8b-Q8_0.gguf LFS Q8	7.95 GB	Download

📊 Model Information

🆔 Model ID: dphn/dolphin-2.9.4-llama3.1-8b-gguf

📅 Created: 2 years ago

🔄 Last Updated: 2 years ago

📥 Downloads: 3.1K

❤️ Likes: 73

🎯 Difficulty: Advanced

⚙️ Quantization: Q2, Q3, Q4, Q5, Q6, Q8

🏷️ Tags

ggufgenerated_from_trainerdataset:cognitivecomputations/Dolphin-2.9dataset:m-a-p/CodeFeedback-Filtered-Instructiondataset:cognitivecomputations/dolphin-coderdataset:cognitivecomputations/samantha-datadataset:microsoft/orca-math-word-problems-200kdataset:mlabonne/FineTome-100kdataset:arcee/agent_datadataset:PawanKrd/math-gpt-4o-200kdataset:cognitivecomputations/SystemChat-2.0base_model:meta-llama/Llama-3.1-8Bbase_model:quantized:meta-llama/Llama-3.1-8Blicense:llama3.1endpoints_compatibleregion:usconversational

🔗 Related Links

🤗 Visit HuggingFace ⚡ Quick Download

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
leaderboard	N/A	none	0	acc	↑	0.2926	±	0.0041
		none	0	acc_norm	↑	0.4513	±	0.0053
		none	0	exact_match	↑	0.0982	±	0.0079
		none	0	instlevelloose_acc	↑	0.3825	±	N/A
		none	0	instlevelstrict_acc	↑	0.3597	±	N/A
		none	0	promptlevelloose_acc	↑	0.2421	±	0.0184
		none	0	promptlevelstrict_acc	↑	0.2181	±	0.0178
- leaderboardbbh	N/A	none	3	accnorm	↑	0.4931	±	0.0061
- leaderboardbbhbooleanexpressions	0	none	3	accnorm	↑	0.8000	±	0.0253
- leaderboardbbhcausaljudgement	0	none	3	accnorm	↑	0.5615	±	0.0364
- leaderboardbbhdateunderstanding	0	none	3	accnorm	↑	0.4520	±	0.0315
- leaderboardbbhdisambiguationqa	0	none	3	accnorm	↑	0.6640	±	0.0299
- leaderboardbbhformalfallacies	0	none	3	accnorm	↑	0.5600	±	0.0315
- leaderboardbbhgeometricshapes	0	none	3	accnorm	↑	0.3640	±	0.0305
- leaderboardbbhhyperbaton	0	none	3	acc_norm	↑	0.6320	±	0.0306
- leaderboardbbhlogicaldeductionfiveobjects	0	none	3	accnorm	↑	0.4600	±	0.0316
- leaderboardbbhlogicaldeductionsevenobjects	0	none	3	accnorm	↑	0.4360	±	0.0314
- leaderboardbbhlogicaldeductionthreeobjects	0	none	3	accnorm	↑	0.6160	±	0.0308
- leaderboardbbhmovierecommendation	0	none	3	accnorm	↑	0.7880	±	0.0259
- leaderboardbbhnavigate	0	none	3	acc_norm	↑	0.5200	±	0.0317
- leaderboardbbhobjectcounting	0	none	3	accnorm	↑	0.4520	±	0.0315
- leaderboardbbhpenguinsinatable	0	none	3	accnorm	↑	0.5205	±	0.0415
- leaderboardbbhreasoningaboutcoloredobjects	0	none	3	accnorm	↑	0.5120	±	0.0317
- leaderboardbbhruinnames	0	none	3	accnorm	↑	0.6320	±	0.0306
- leaderboardbbhsalienttranslationerrordetection	0	none	3	accnorm	↑	0.4320	±	0.0314
- leaderboardbbhsnarks	0	none	3	acc_norm	↑	0.5843	±	0.0370
- leaderboardbbhsportsunderstanding	0	none	3	accnorm	↑	0.7040	±	0.0289
- leaderboardbbhtemporalsequences	0	none	3	accnorm	↑	0.1440	±	0.0222
- leaderboardbbhtrackingshuffledobjectsfiveobjects	0	none	3	acc_norm	↑	0.1560	±	0.0230
- leaderboardbbhtrackingshuffledobjectssevenobjects	0	none	3	acc_norm	↑	0.1320	±	0.0215
- leaderboardbbhtrackingshuffledobjectsthreeobjects	0	none	3	acc_norm	↑	0.2840	±	0.0286
- leaderboardbbhweboflies	0	none	3	acc_norm	↑	0.4840	±	0.0317
- leaderboardgpqa	N/A	none	0	accnorm	↑	0.2903	±	0.0132
- leaderboardgpqadiamond	1	none	0	acc_norm	↑	0.2980	±	0.0326
- leaderboardgpqaextended	1	none	0	acc_norm	↑	0.2839	±	0.0193
- leaderboardgpqamain	1	none	0	acc_norm	↑	0.2946	±	0.0216
- leaderboardifeval	2	none	0	instlevellooseacc	↑	0.3825	±	N/A
		none	0	instlevelstrict_acc	↑	0.3597	±	N/A
		none	0	promptlevelloose_acc	↑	0.2421	±	0.0184
		none	0	promptlevelstrict_acc	↑	0.2181	±	0.0178
- leaderboardmathalgebrahard	1	none	4	exactmatch	↑	0.1596	±	0.0209
- leaderboardmathcountingandprobhard	1	none	4	exactmatch	↑	0.0488	±	0.0195
- leaderboardmathgeometryhard	1	none	4	exactmatch	↑	0.0530	±	0.0196
- leaderboardmathhard	N/A	none	4	exact_match	↑	0.0982	±	0.0079
- leaderboardmathintermediatealgebrahard	1	none	4	exact_match	↑	0.0143	±	0.0071
- leaderboardmathnumtheoryhard	1	none	4	exact_match	↑	0.0455	±	0.0168
- leaderboardmathprealgebrahard	1	none	4	exactmatch	↑	0.2591	±	0.0316
- leaderboardmathprecalculushard	1	none	4	exactmatch	↑	0.0519	±	0.0192
- leaderboardmmlupro	0.1	none	5	acc	↑	0.2926	±	0.0041
- leaderboardmusr	N/A	none	0	accnorm	↑	0.3862	±	0.0173
- leaderboardmusrmurdermysteries	1	none	0	accnorm	↑	0.5280	±	0.0316
- leaderboardmusrobjectplacements	1	none	0	accnorm	↑	0.3594	±	0.0300
- leaderboardmusrteamallocation	1	none	0	accnorm	↑	0.2720	±	0.0282