38 Pulls Updated 7 weeks ago
Readme
Dolphin 2.9.3 Mistral 7b v0.3 32k 🐬
Curated and trained by Eric Hartford and Cognitive Computations
Discord: https://discord.gg/h3K4XGj2RH
Our appreciation for the sponsors of Dolphin 2.9.3:
- Crusoe Cloud - provided excellent on-demand 8xH100 node
- OnDemand - provided inference sponsorship
This model is based on mistralai/Mistral-7B-v0.3, and is governed by the apache 2.0 license.
The base model has 32k context, and our finetuning took place with 8192 sequence length.
Dolphin 2.9.3 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-2.9.3 has a variety of instruction following, conversational, and coding skills. It also has initial agentic abilities and supports function calling.
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.
Dolphin is licensed according to apache 2.0 license. We grant permission for any use, including commercial. Dolphin was trained on data generated from GPT4, among other models.
Evals
https://huggingface.co/cognitivecomputations/dolphin-2.9.3-mistral-7B-32k
Training
See axolotl config
axolotl version: 0.4.0
”`yaml
base_model: mistralai/Mistral-7B-v0.3
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: true
strict: false
datasets:
- path: /workspace/datasets/dolphin-2.9.3/dolphin201-sharegpt2.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/SystemChat_filtered_sharegpt.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/SystemChat_multilingual_sharegpt.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/dolphin-coder-translate-sharegpt2.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/dolphin-coder-codegen-sharegpt2.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/m-a-p_Code-Feedback-sharegpt-unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/m-a-p_CodeFeedback-Filtered-Instruction-sharegpt-unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/not_samantha_norefusals.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/Orca-Math-resort-unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/agent_instruct_react_unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/toolbench_instruct_j1s1_3k_unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/toolbench_negative_unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/toolbench_react_10p_unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/toolbench_tflan_cot_30p_unfiltered.jsonl
type: sharegpt
conversation: chatml
- path: /workspace/datasets/dolphin-2.9.3/openhermes200k_unfiltered.jsonl
type: sharegpt
conversation: chatml
chat_template: chatml
adapter: qlora
lora_r: 128
lora_alpha: 16
lora_modules_to_save: [embed_tokens, lm_head]
lora_dropout: 0.05
lora_target_linear: true
dataset_prepared_path: /workspace/axolotl/dolph-2.9.3-prepared
val_set_size: 0.01
output_dir: /workspace/axolotl/dolphin-2.9.3-mistral-7B
sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true
wandb_project: dolphin-2.9.3-Mistral-7B
wandb_watch:
wandb_run_id:
wandb_log_model:
gradient_accumulation_steps: 16
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_8bit
lr_scheduler: cosine
learning_rate: 5e-6
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32:
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 100
evals_per_epoch: 4
eval_table_size:
saves_per_epoch: 1
save_total_limit: 2
save_steps:
debug:
deepspeed: /workspace/axolotl/deepspeed_configs/zero3_bf16.json
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:
eos_token: “<|im_end|>”
tokens:
- “<|im_start|>”