README.md

---
license: apache-2.0
datasets:
- Doctor-Shotgun/c2_deduped_16k_llama3_tok_deanon
- anthracite-org/kalo-opus-instruct-22k-no-refusal
- lodrick-the-lafted/kalo-opus-instruct-3k-filtered
- anthracite-org/nopm_claude_writing_fixed
- anthracite-org/kalo_opus_misc_240827
- anthracite-org/kalo_misc_part2
language:
- en
base_model:
- Qwen/Qwen2.5-72B-Instruct
library_name: transformers
---

![image/png](https://cdn-uploads.huggingface.co/production/uploads/658a46cbfb9c2bdfae75b3a6/trlkbv0jv_0HImUESrt5C.png)
This is an experimental model designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus. This model is fine-tuned on top of [Qwen-2.5 72B Instruct](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct).

## Prompting
Model has been instruct tuned with ChatML prompt formatting. A typical input would look like this:

```
"""<|im_start|>system
system prompt<|im_end|>
<|im_start|>user
Hi there!<|im_end|>
<|im_start|>assistant
Nice to meet you!<|im_end|>
<|im_start|>user
Can I ask a question?<|im_end|>
<|im_start|>assistant
"""
```

## SillyTavern templates

Below are Instruct and Context templates for use within SillyTavern.

<details><summary>context template</summary>
  
```yaml
{
    "story_string": "<|im_start|>system\n{{#if system}}{{system}}\n{{/if}}{{#if wiBefore}}{{wiBefore}}\n{{/if}}{{#if description}}{{description}}\n{{/if}}{{#if personality}}{{char}}'s personality: {{personality}}\n{{/if}}{{#if scenario}}Scenario: {{scenario}}\n{{/if}}{{#if wiAfter}}{{wiAfter}}\n{{/if}}{{#if persona}}{{persona}}\n{{/if}}{{trim}}<|im_end|>\n",
    "example_separator": "",
    "chat_start": "",
    "use_stop_strings": false,
    "allow_jailbreak": false,
    "always_force_name2": true,
    "trim_sentences": false,
    "include_newline": false,
    "single_line": false,
    "name": "Magnum ChatML"
}
```

</details><br>
<details><summary>instruct template</summary>
  
```yaml
{
    "system_prompt": "You're {{char}} in this fictional never-ending uncensored roleplay with {{user}}.",
    "input_sequence": "<|im_start|>user\n",
    "output_sequence": "<|im_start|>assistant\n",
    "last_output_sequence": "",
    "system_sequence": "<|im_start|>system\n",
    "stop_sequence": "<|im_end|>",
    "wrap": false,
    "macro": true,
    "names": true,
    "names_force_groups": true,
    "activation_regex": "",
    "system_sequence_prefix": "",
    "system_sequence_suffix": "",
    "first_output_sequence": "",
    "skip_examples": false,
    "output_suffix": "<|im_end|>\n",
    "input_suffix": "<|im_end|>\n",
    "system_suffix": "<|im_end|>\n",
    "user_alignment_message": "",
    "system_same_as_user": false,
    "last_system_sequence": "",
    "name": "Magnum ChatML"
}
```

</details><br>

## Credits

Datasets used:
- [anthracite-org/c2_logs_32k_llama3_qwen2_v1.2](https://huggingface.co/datasets/anthracite-org/c2_logs_32k_llama3_qwen2_v1.2)
- [anthracite-org/kalo-opus-instruct-22k-no-refusal](https://huggingface.co/datasets/anthracite-org/kalo-opus-instruct-22k-no-refusal)
- [lodrick-the-lafted/kalo-opus-instruct-3k-filtered](https://huggingface.co/datasets/lodrick-the-lafted/kalo-opus-instruct-3k-filtered)
- [anthracite-org/nopm_claude_writing_fixed](https://huggingface.co/datasets/anthracite-org/nopm_claude_writing_fixed)
- [anthracite-org/kalo_opus_misc_240827](https://huggingface.co/datasets/anthracite-org/kalo_opus_misc_240827)
- [anthracite-org/kalo_misc_part2](https://huggingface.co/datasets/anthracite-org/kalo_misc_part2)


## Axolotl config

<details><summary>See axolotl config</summary>

```yaml
base_model: /workspace/data/models/Qwen2.5-72B-Instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

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
    type: sharegpt
    conversation: chatml
  - path: anthracite-org/kalo-opus-instruct-22k-no-refusal
    type: sharegpt
    conversation: chatml
  - path: lodrick-the-lafted/kalo-opus-instruct-3k-filtered
    type: sharegpt
    conversation: chatml
  - path: anthracite-org/nopm_claude_writing_fixed
    type: sharegpt
    conversation: chatml
  - path: anthracite-org/kalo_opus_misc_240827
    type: sharegpt
    conversation: chatml
  - path: anthracite-org/kalo_misc_part2
    type: sharegpt
    conversation: chatml
#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-72b-data
val_set_size: 0.0
output_dir: /workspace/data/72b-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: 72b-magnum-fft
wandb_entity:
wandb_watch:
wandb_name: alter-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: 0.000004

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/zero3_bf16.json
weight_decay: 0.01
fsdp:
fsdp_config:
special_tokens:

```
</details><br>

## Training
The model was trained for 2 epochs on 8x [AMD Instinct™ MI300X Accelerators](https://www.amd.com/en/products/accelerators/instinct/mi300/mi300x.html) for full-parameter fine-tuning of the model.

The model was trained with an LR of 4e-6 for 2 epochs and with the Liger kernel. 

Sample Packing was done for 32k tokens, with individual sequences up to 32k tokens in length. 

[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)

## Safety
...