This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus.

This model is fine-tuned on top of mistralai/Mistral-Nemo-Instruct-2407.

Prompting

A typical input would look like this:

<s>[INST] SYSTEM MESSAGE
USER MESSAGE[/INST] ASSISTANT MESSAGE</s>[INST] USER MESSAGE[/INST]

SillyTavern templates

Below are Instruct and Context templates for use within SillyTavern.

default SillyTavern template works fine

default SillyTavern template works fine

Axolotl config

base_model: mistralai/Mistral-Nemo-Instruct-2407
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

hub_model_id: anthracite-org/magnum-v4-12b-r2
hub_strategy: "all_checkpoints"
push_dataset_to_hub:
hf_use_auth_token: true

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_swiglu: true
liger_fused_linear_cross_entropy: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: anthracite-org/c2_logs_32k_llama3_qwen2_v1.2_no_system
    type: custommistralv3tekken
  - path: anthracite-org/kalo-opus-instruct-22k-no-refusal-no-system
    type: custommistralv3tekken
  - path: anthracite-org/kalo-opus-instruct-3k-filtered-no-system
    type: custommistralv3tekken
  - path: anthracite-org/nopm_claude_writing_fixed
    type: custommistralv3tekken
  - path: anthracite-org/kalo_opus_misc_240827_no_system
    type: custommistralv3tekken
  - path: anthracite-org/kalo_misc_part2_no_system
    type: custommistralv3tekken
#chat_template: chatml
shuffle_merged_datasets: true
#default_system_message: "You are an assistant that responds to the user."
dataset_prepared_path: /workspace/data/magnum-12b-data
val_set_size: 0.0
output_dir: /workspace/data/12b-fft-out

sequence_len: 32768
sample_packing: true
pad_to_sequence_len: true

adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:

wandb_project: 12b-magnum-fft
wandb_entity:
wandb_watch:
wandb_name: v4-r2-attempt-01
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.00001

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

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

warmup_steps: 40
evals_per_epoch:
eval_table_size:
eval_max_new_tokens:
saves_per_epoch: 2
debug:
deepspeed: deepspeed_configs/zero2.json
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:
  pad_token: <pad>

Credits

We'd like to thank Recursal / Featherless for sponsoring the compute for this train, Featherless has been hosting our Magnum models since the first 72 B and has given thousands of people access to our models and helped us grow.

We would also like to thank all members of Anthracite who made this finetune possible.

Datasets

• anthracite-org/c2_logs_32k_llama3_qwen2_v1.2_no_system
• anthracite-org/kalo-opus-instruct-22k-no-refusal-no-system
• anthracite-org/kalo-opus-instruct-3k-filtered-no-system
• anthracite-org/nopm_claude_writing_fixed
• anthracite-org/kalo_opus_misc_240827_no_system
• anthracite-org/kalo_misc_part2_no_system

Training

The training was done for 2 epochs. We used 8xH100s GPUs graciously provided by Recursal AI / Featherless AI for the full-parameter fine-tuning of the model.

Safety

...

Open LLM Leaderboard Evaluation Results

Detailed results can be found here