AlphaMonarch-laser
AlphaMonarch-laser is a new DPO merge using laserQLoRA that retains all the reasoning abilities of the very best merges and significantly improves its conversational abilities. Kind of the best of both worlds in a 7B model. This model uses mlabonne/NeuralMonarch-7B as its base model, finetuned on only half of the layers using laserQLoRA. The preference dataset used for DPO is mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha.
Evaluation data
Task Version Metric Value StdErr
agieval_aqua_rat 0 acc 28.35% 2.83% agieval_aqua_rat 0 acc_norm 26.38% 2.77% agieval_logiqa_en 0 acc 38.25% 1.91% agieval_logiqa_en 0 acc_norm 38.10% 1.90% agieval_lsat_ar 0 acc 23.91% 2.82% agieval_lsat_ar 0 acc_norm 23.48% 2.80% agieval_lsat_lr 0 acc 52.75% 2.21% agieval_lsat_lr 0 acc_norm 53.92% 2.21% agieval_lsat_rc 0 acc 66.91% 2.87% agieval_lsat_rc 0 acc_norm 67.29% 2.87% agieval_sat_en 0 acc 78.64% 2.86% agieval_sat_en 0 acc_norm 78.64% 2.86% agieval_sat_en_without_passage 0 acc 45.15% 3.48% agieval_sat_en_without_passage 0 acc_norm 44.17% 3.47% agieval_sat_math 0 acc 33.18% 3.18% agieval_sat_math 0 acc_norm 31.36% 3.14%
🏆 Evaluation
| Task | Version | Metric | Value | StdErr |
|---|---|---|---|---|
| agieval_aqua_rat | 0 | acc | 28.35% | 2.83% |
| agieval_aqua_rat | 0 | acc_norm | 26.38% | 2.77% |
| agieval_logiqa_en | 0 | acc | 38.25% | 1.91% |
| agieval_logiqa_en | 0 | acc_norm | 38.10% | 1.90% |
| agieval_lsat_ar | 0 | acc | 23.91% | 2.82% |
| agieval_lsat_ar | 0 | acc_norm | 23.48% | 2.80% |
| agieval_lsat_lr | 0 | acc | 52.75% | 2.21% |
| agieval_lsat_lr | 0 | acc_norm | 53.92% | 2.21% |
| agieval_lsat_rc | 0 | acc | 66.91% | 2.87% |
| agieval_lsat_rc | 0 | acc_norm | 67.29% | 2.87% |
| agieval_sat_en | 0 | acc | 78.64% | 2.86% |
| agieval_sat_en | 0 | acc_norm | 78.64% | 2.86% |
| agieval_sat_en_without_passage | 0 | acc | 45.15% | 3.48% |
| agieval_sat_en_without_passage | 0 | acc_norm | 44.17% | 3.47% |
| agieval_sat_math | 0 | acc | 33.18% | 3.18% |
| agieval_sat_math | 0 | acc_norm | 31.36% | 3.14% |
Average: 75.9% without mmlu
TruthfulQA
| Task | Version | Metric | Value | Stderr | |
|---|---|---|---|---|---|
| truthfulqa_mc | 1 | mc1 | 63.03 | ± | 1.68 |
| mc2 | 78.39 | ± | 1.37 |
BigBench Reasoning Test
| Task | Version | Metric | Value | Stderr | |
|---|---|---|---|---|---|
| bigbench_causal_judgement | 0 | multiple_choice_grade | 60.00 | _ | 3.56 |
| bigbench_date_understanding | 0 | multiple_choice_grade | 62.06 | _ | 2.53 |
| bigbench_disambiguation_qa | 0 | multiple_choice_grade | 54.26 | _ | 3.11 |
| bigbench_geometric_shapes | 0 | multiple_choice_grade | 23.96 | _ | 2.26 |
| ... | exact_str_match | ||||
| bigbench_geometric_shapes | 0 | exact_str_match | 0.00 | _ | 0.00 |
| bigbench_logical_deduction_five_objects | 0 | multiple_choice_grade | 32.80 | _ | 2.10 |
| bigbench_logical_deduction_seven_objects | 0 | multiple_choice_grade | 23.86 | _ | 1.61 |
| bigbench_logical_deduction_three_objects | 0 | multiple_choice_grade | 59.33 | _ | 2.84 |
| bigbench_movie_recommendation | 0 | multiple_choice_grade | 58.00 | _ | 2.21 |
| bigbench_navigate | 0 | multiple_choice_grade | 56.00 | _ | 1.57 |
| bigbench_reasoning_about_colored_objects | 0 | multiple_choice_grade | 69.20 | _ | 1.03 |
| bigbench_ruin_names | 0 | multiple_choice_grade | 55.36 | _ | 2.35 |
| bigbench_salient_translation_error_detection | 0 | multiple_choice_grade | 41.48 | _ | 1.56 |
| bigbench_snarks | 0 | multiple_choice_grade | 73.48 | _ | 3.29 |
| bigbench_sports_understanding | 0 | multiple_choice_grade | 76.06 | _ | 1.36 |
| bigbench_temporal_sequences | 0 | multiple_choice_grade | 55.50 | _ | 1.57 |
| bigbench_tracking_shuffled_objects_five_objects | 0 | multiple_choice_grade | 23.28 | _ | 1.20 |
| bigbench_tracking_shuffled_objects_seven_objects | 0 | multiple_choice_grade | 19.37 | _ | 0.94 |
| bigbench_tracking_shuffled_objects_three_objects | 0 | multiple_choice_grade | 59.33 | _ | 2.84 |
Average: 49.08%
GPT4ALL
| Task | Version | Metric | Value | Stderr | |
|---|---|---|---|---|---|
| arc_challenge | 0 | acc | 66.29 | _ | 1.38 |
| acc_norm | 68.26 | _ | 1.36 | ||
| arc_easy | 0 | acc | 86.57 | _ | 0.70 |
| acc_norm | 80.81 | _ | 0.81 | ||
| boolq | 1 | acc | 87.16 | _ | 0.59 |
| hellaswag | 0 | acc | 69.60 | _ | 0.46 |
| acc_norm | 87.45 | _ | 0.33 | ||
| openbookqa | 0 | acc | 39.20 | _ | 2.19 |
| acc_norm | 49.60 | _ | 2.24 | ||
| piqa | 0 | acc | 83.03 | _ | 0.88 |
| acc_norm | 84.87 | _ | 0.84 | ||
| winogrande | 0 | acc | 81.06 | _ | 1.10 |
Average: 68.75%
AGIEVAL
Here is the converted table in the required format, including multiplication of all values by 100 and calculating the average for the value column:
| Task | Version | Metric | Value | StdErr |
|---|---|---|---|---|
| agieval_aqua_rat | 0 | acc | 28.35 | 2.83 |
| acc_norm | 26.38 | 2.77 | ||
| agieval_logiqa_en | 0 | acc | 38.25 | 1.91 |
| acc_norm | 38.09 | 1.90 | ||
| agieval_lsat_ar | 0 | acc | 23.91 | 2.82 |
| acc_norm | 23.48 | 2.80 | ||
| agieval_lsat_lr | 0 | acc | 52.75 | 2.21 |
| acc_norm | 53.92 | 2.21 | ||
| agieval_lsat_rc | 0 | acc | 66.91 | 2.87 |
| acc_norm | 67.29 | 2.87 | ||
| agieval_sat_en | 0 | acc | 78.64 | 2.86 |
| acc_norm | 78.64 | 2.86 | ||
| agieval_sat_en_without_passage | 0 | acc | 45.15 | 3.48 |
| acc_norm | 44.17 | 3.47 | ||
| agieval_sat_math | 0 | acc | 33.18 | 3.18 |
| acc_norm | 31.36 | 3.14 |
Average: 47.44%
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-07
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 1080
📝 Axolotl Configuration
base_model: mlabonne/NeuralMonarch-7B
model_type: MistralForCausalLM
tokenizer_type: LlamaTokenizer
is_mistral_derived_model: true
load_in_8bit: false
load_in_4bit: true
strict: false
rl: dpo
chat_template: chatml
datasets:
- path: mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha
split: train
type: chatml.intel
dataset_prepared_path:
val_set_size: 0.01
output_dir: ./out
adapter: qlora
lora_model_dir:
sequence_len: 1800
sample_packing: false
pad_to_sequence_len: false
lora_r: 16
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:
- layers.1.self_attn.q_proj
- layers.0.self_attn.q_proj
- layers.15.self_attn.q_proj
- layers.12.self_attn.q_proj
- layers.11.self_attn.q_proj
- layers.14.self_attn.q_proj
- layers.9.self_attn.q_proj
- layers.16.self_attn.q_proj
- layers.30.self_attn.q_proj
- layers.18.self_attn.q_proj
- layers.13.self_attn.q_proj
- layers.10.self_attn.q_proj
- layers.7.self_attn.q_proj
- layers.8.self_attn.q_proj
- layers.4.self_attn.q_proj
- layers.19.self_attn.q_proj
- layers.27.self_attn.k_proj
- layers.24.self_attn.k_proj
- layers.25.self_attn.k_proj
- layers.22.self_attn.k_proj
- layers.26.self_attn.k_proj
- layers.29.self_attn.k_proj
- layers.23.self_attn.k_proj
- layers.28.self_attn.k_proj
- layers.21.self_attn.k_proj
- layers.31.self_attn.k_proj
- layers.30.self_attn.k_proj
- layers.20.self_attn.k_proj
- layers.5.self_attn.k_proj
- layers.19.self_attn.k_proj
- layers.17.self_attn.k_proj
- layers.18.self_attn.k_proj
- layers.19.self_attn.v_proj
- layers.24.self_attn.v_proj
- layers.18.self_attn.v_proj
- layers.5.self_attn.v_proj
- layers.3.self_attn.v_proj
- layers.16.self_attn.v_proj
- layers.23.self_attn.v_proj
- layers.27.self_attn.v_proj
- layers.25.self_attn.v_proj
- layers.26.self_attn.v_proj
- layers.20.self_attn.v_proj
- layers.6.self_attn.v_proj
- layers.15.self_attn.v_proj
- layers.17.self_attn.v_proj
- layers.29.self_attn.v_proj
- layers.22.self_attn.v_proj
- layers.12.self_attn.o_proj
- layers.9.self_attn.o_proj
- layers.14.self_attn.o_proj
- layers.0.self_attn.o_proj
- layers.6.self_attn.o_proj
- layers.8.self_attn.o_proj
- layers.10.self_attn.o_proj
- layers.11.self_attn.o_proj
- layers.13.self_attn.o_proj
- layers.24.self_attn.o_proj
- layers.7.self_attn.o_proj
- layers.15.self_attn.o_proj
- layers.5.self_attn.o_proj
- layers.17.self_attn.o_proj
- layers.25.self_attn.o_proj
- layers.4.self_attn.o_proj
- layers.31.mlp.gate_proj
- layers.30.mlp.gate_proj
- layers.4.mlp.gate_proj
- layers.3.mlp.gate_proj
- layers.29.mlp.gate_proj
- layers.28.mlp.gate_proj
- layers.6.mlp.gate_proj
- layers.27.mlp.gate_proj
- layers.5.mlp.gate_proj
- layers.26.mlp.gate_proj
- layers.25.mlp.gate_proj
- layers.7.mlp.gate_proj
- layers.2.mlp.gate_proj
- layers.24.mlp.gate_proj
- layers.23.mlp.gate_proj
- layers.10.mlp.gate_proj
- layers.6.mlp.up_proj
- layers.4.mlp.up_proj
- layers.5.mlp.up_proj
- layers.27.mlp.up_proj
- layers.25.mlp.up_proj
- layers.26.mlp.up_proj
- layers.17.mlp.up_proj
- layers.24.mlp.up_proj
- layers.7.mlp.up_proj
- layers.10.mlp.up_proj
- layers.3.mlp.up_proj
- layers.11.mlp.up_proj
- layers.23.mlp.up_proj
- layers.9.mlp.up_proj
- layers.14.mlp.up_proj
- layers.18.mlp.up_proj
- layers.19.mlp.down_proj
- layers.20.mlp.down_proj
- layers.18.mlp.down_proj
- layers.21.mlp.down_proj
- layers.29.mlp.down_proj
- layers.1.mlp.down_proj
- layers.22.mlp.down_proj
- layers.28.mlp.down_proj
- layers.23.mlp.down_proj
- layers.30.mlp.down_proj
- layers.17.mlp.down_proj
- layers.4.mlp.down_proj
- layers.2.mlp.down_proj
- layers.15.mlp.down_proj
- layers.5.mlp.down_proj
wandb_project: axolotl
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
gradient_accumulation_steps: 8
micro_batch_size: 1
num_epochs: 1
optimizer: paged_adamw_32bit
lr_scheduler: cosine
learning_rate: 5e-7
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: true
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 100
evals_per_epoch: 1
eval_table_size:
eval_table_max_new_tokens: 128
save_steps: 1080
max_steps: 1080
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.2+cu118
- Datasets 2.17.0
- Tokenizers 0.15.0
- axolotl: 0.4.0
- Downloads last month
- 0