shisa-ai/shisa-v1-swallowmx-13a47b

shisa-v2 Base Model ablation

Using a fork of Lightblue's Shaberi benchmark framework:

Model	Average	ELYZA-tasks-100	MT-Bench	Rakuda	Tengu-Bench
gpt-4-turbo-2024-04-09	8.75	8.78	8.74	9.18	8.31
CohereForAI/c4ai-command-r-plus	7.69	7.50	7.43	9.05	6.79
gpt-3.5-turbo-0125	7.17	7.24	6.98	7.64	6.82
shisa-ai/shisa-v1-llama3-70b	7.17	7.16	7.45	7.98	6.09
karakuri-ai/karakuri-lm-70b-chat-v0.1	6.84	6.86	6.43	7.85	6.23
lightblue/ao-karasu-72B	6.81	7.19	6.54	7.25	6.27
shisa-ai/shisa-v1-llama3-8b^	6.29	6.62	6.41	7.05	5.07
shisa-ai/shisa-swallowmx-13a47b-v1	6.17	6.48	6.07	7.11	5.03
shisa-ai/shisa-v1-llama3-8b	6.10	6.52	6.20	6.37	5.33
Rakuten/RakutenAI-7B-chat	5.58	5.92	4.60	6.58	5.24
shisa-ai/shisa-v1-gemma-8b	5.64	6.50	5.42	5.10	5.55
augmxnt/shisa-gamma-7b-v1	5.56	5.84	4.00	6.73	5.68
lightblue/qarasu-14B-chat-plus-unleashed	5.20	5.58	4.74	5.46	5.01
cyberagent/calm2-7b-chat	4.76	4.90	3.58	5.75	4.81
mistralai/Mistral-7B-Instruct-v0.2	4.69	5.78	4.65	3.80	4.53
shisa-ai/shisa-v1-yi1.5-9b	4.63	5.98	4.28	3.26	5.00

^ Sampler settings: temperature 0.2, min_p 0.1, frequency_penalty 0.5

See axolotl config

axolotl version: 0.4.0

base_model: tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1
model_type: AutoModelForCausalLM
tokenizer_type: LlamaTokenizer
trust_remote_code: true

load_in_8bit: false
load_in_4bit: false
strict: false

chat_template: inst
datasets:
  - path: augmxnt/ultra-orca-boros-en-ja-v1
    type: sharegpt
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/basemodel-swallowmx-8x22b

model_config:
  output_router_logits: true

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true

use_wandb: true
wandb_project: shisa-v2
wandb_entity: augmxnt
wandb_name: shisa-swallowmx-13a47b-v1

global_batch_size: 1
gradient_accumulation_steps: 8
micro_batch_size: 1
num_epochs: 3
# https://github.com/huggingface/transformers/issues/22101
# https://github.com/huggingface/transformers/blob/main/src/transformers/training_args.py#L141
optimizer: paged_adamw_8bit
lr_scheduler: linear
learning_rate: 2e-5

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.1
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed: axolotl/deepspeed_configs/zero3_bf16.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

outputs/basemodel-swallowmx-8x22b

This model is a fine-tuned version of tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4443

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:

• learning_rate: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 119
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.5705	0.0022	1	0.5065
0.505	0.4993	229	0.3910
0.5258	0.9986	458	0.3654
0.2964	1.4835	687	0.3786
0.2923	1.9828	916	0.3669
0.1462	2.4682	1145	0.4429
0.1156	2.9676	1374	0.4443

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1