gemma-instruct-merge-named_correctly

gemma-instruct-merge-named_correctly is a merge of the following models using LazyMergekit:

• google/gemma-2-2b-it

🧩 Configuration

models:
- model: google/gemma-2-2b
- model: google/gemma-2-2b-it
  parameters:
    density:
    - filter: model.layers.0.self_attn.q_proj
      value: 0.00539
    - filter: model.layers.1.self_attn.q_proj
      value: 0.3256
    - filter: model.layers.2.self_attn.q_proj
      value: 0.03843
    - filter: model.layers.3.self_attn.q_proj
      value: 0.14886
    - filter: model.layers.4.self_attn.q_proj
      value: 0.41323
    - filter: model.layers.5.self_attn.q_proj
      value: 0.33564
    - filter: model.layers.6.self_attn.q_proj
      value: 0.03716
    - filter: model.layers.7.self_attn.q_proj
      value: 0.1084
    - filter: model.layers.8.self_attn.q_proj
      value: 0.16658
    - filter: model.layers.9.self_attn.q_proj
      value: 0.19716
    - filter: model.layers.10.self_attn.q_proj
      value: 0.10008
    - filter: model.layers.11.self_attn.q_proj
      value: 0.37198
    - filter: model.layers.12.self_attn.q_proj
      value: 0.3622
    - filter: model.layers.13.self_attn.q_proj
      value: 0.26806
    - filter: model.layers.14.self_attn.q_proj
      value: 0.73192
    - filter: model.layers.15.self_attn.q_proj
      value: 0.14731
    - filter: model.layers.16.self_attn.q_proj
      value: 0.46923
    - filter: model.layers.17.self_attn.q_proj
      value: 0.63063
    - filter: model.layers.18.self_attn.q_proj
      value: 0.65673
    - filter: model.layers.19.self_attn.q_proj
      value: 0.66558
    - filter: model.layers.20.self_attn.q_proj
      value: 0.63758
    - filter: model.layers.21.self_attn.q_proj
      value: 0.64362
    - filter: model.layers.22.self_attn.q_proj
      value: 0.40341
    - filter: model.layers.23.self_attn.q_proj
      value: 0.12982
    - filter: model.layers.24.self_attn.q_proj
      value: 0.04552
    - filter: model.layers.25.self_attn.q_proj
      value: 0.03919
    - filter: model.layers.0.self_attn.k_proj
      value: 0.00592
    - filter: model.layers.1.self_attn.k_proj
      value: 0.18684
    - filter: model.layers.2.self_attn.k_proj
      value: 0.02603
    - filter: model.layers.3.self_attn.k_proj
      value: 0.07283
    - filter: model.layers.4.self_attn.k_proj
      value: 0.27067
    - filter: model.layers.5.self_attn.k_proj
      value: 0.18189
    - filter: model.layers.6.self_attn.k_proj
      value: 0.10695
    - filter: model.layers.7.self_attn.k_proj
      value: 0.17169
    - filter: model.layers.8.self_attn.k_proj
      value: 0.08753
    - filter: model.layers.9.self_attn.k_proj
      value: 0.12807
    - filter: model.layers.10.self_attn.k_proj
      value: 0.07783
    - filter: model.layers.11.self_attn.k_proj
      value: 0.41518
    - filter: model.layers.12.self_attn.k_proj
      value: 0.42307
    - filter: model.layers.13.self_attn.k_proj
      value: 0.2896
    - filter: model.layers.14.self_attn.k_proj
      value: 0.52953
    - filter: model.layers.15.self_attn.k_proj
      value: 0.25035
    - filter: model.layers.16.self_attn.k_proj
      value: 0.57679
    - filter: model.layers.17.self_attn.k_proj
      value: 0.52768
    - filter: model.layers.18.self_attn.k_proj
      value: 0.666
    - filter: model.layers.19.self_attn.k_proj
      value: 0.67796
    - filter: model.layers.20.self_attn.k_proj
      value: 0.58825
    - filter: model.layers.21.self_attn.k_proj
      value: 0.51447
    - filter: model.layers.22.self_attn.k_proj
      value: 0.33317
    - filter: model.layers.23.self_attn.k_proj
      value: 0.05987
    - filter: model.layers.24.self_attn.k_proj
      value: 0.02903
    - filter: model.layers.25.self_attn.k_proj
      value: 0.08715
    - filter: model.layers.0.self_attn.v_proj
      value: 0.03025
    - filter: model.layers.1.self_attn.v_proj
      value: 0.14137
    - filter: model.layers.2.self_attn.v_proj
      value: 0.00286
    - filter: model.layers.3.self_attn.v_proj
      value: 0.09155
    - filter: model.layers.4.self_attn.v_proj
      value: 0.25172
    - filter: model.layers.5.self_attn.v_proj
      value: 0.2062
    - filter: model.layers.6.self_attn.v_proj
      value: 0.06811
    - filter: model.layers.7.self_attn.v_proj
      value: 0.01334
    - filter: model.layers.8.self_attn.v_proj
      value: 0.18251
    - filter: model.layers.9.self_attn.v_proj
      value: 0.10415
    - filter: model.layers.10.self_attn.v_proj
      value: 0.04
    - filter: model.layers.11.self_attn.v_proj
      value: 0.22664
    - filter: model.layers.12.self_attn.v_proj
      value: 0.16337
    - filter: model.layers.13.self_attn.v_proj
      value: 0.09347
    - filter: model.layers.14.self_attn.v_proj
      value: 0.52549
    - filter: model.layers.15.self_attn.v_proj
      value: 0.02152
    - filter: model.layers.16.self_attn.v_proj
      value: 0.25325
    - filter: model.layers.17.self_attn.v_proj
      value: 0.43041
    - filter: model.layers.18.self_attn.v_proj
      value: 0.49113
    - filter: model.layers.19.self_attn.v_proj
      value: 0.32307
    - filter: model.layers.20.self_attn.v_proj
      value: 0.41922
    - filter: model.layers.21.self_attn.v_proj
      value: 0.32166
    - filter: model.layers.22.self_attn.v_proj
      value: 0.34057
    - filter: model.layers.23.self_attn.v_proj
      value: 0.24319
    - filter: model.layers.24.self_attn.v_proj
      value: 0.07956
    - filter: model.layers.25.self_attn.v_proj
      value: 0.13849
    - filter: model.layers.0.self_attn.o_proj
      value: 0.03265
    - filter: model.layers.1.self_attn.o_proj
      value: 0.22645
    - filter: model.layers.2.self_attn.o_proj
      value: 0.06134
    - filter: model.layers.3.self_attn.o_proj
      value: 0.16228
    - filter: model.layers.4.self_attn.o_proj
      value: 0.07924
    - filter: model.layers.5.self_attn.o_proj
      value: 0.1259
    - filter: model.layers.6.self_attn.o_proj
      value: 0.09982
    - filter: model.layers.7.self_attn.o_proj
      value: 0.02826
    - filter: model.layers.8.self_attn.o_proj
      value: 0.03906
    - filter: model.layers.9.self_attn.o_proj
      value: 1.0
    - filter: model.layers.10.self_attn.o_proj
      value: 0.51227
    - filter: model.layers.11.self_attn.o_proj
      value: 0.34671
    - filter: model.layers.12.self_attn.o_proj
      value: 0.70159
    - filter: model.layers.13.self_attn.o_proj
      value: 0.4424
    - filter: model.layers.14.self_attn.o_proj
      value: 0.74354
    - filter: model.layers.15.self_attn.o_proj
      value: 0.16266
    - filter: model.layers.16.self_attn.o_proj
      value: 0.71896
    - filter: model.layers.17.self_attn.o_proj
      value: 0.53179
    - filter: model.layers.18.self_attn.o_proj
      value: 0.18472
    - filter: model.layers.19.self_attn.o_proj
      value: 0.09507
    - filter: model.layers.20.self_attn.o_proj
      value: 0.13472
    - filter: model.layers.21.self_attn.o_proj
      value: 0.23246
    - filter: model.layers.22.self_attn.o_proj
      value: 0.10599
    - filter: model.layers.23.self_attn.o_proj
      value: 0.00282
    - filter: model.layers.24.self_attn.o_proj
      value: 0.09864
    - filter: model.layers.25.self_attn.o_proj
      value: 0.00961
    - filter: model.layers.0.mlp.gate_proj
      value: 0.4928
    - filter: model.layers.1.mlp.gate_proj
      value: 0.08775
    - filter: model.layers.2.mlp.gate_proj
      value: 0.0002
    - filter: model.layers.3.mlp.gate_proj
      value: 0.12885
    - filter: model.layers.4.mlp.gate_proj
      value: 0.50973
    - filter: model.layers.5.mlp.gate_proj
      value: 0.29085
    - filter: model.layers.6.mlp.gate_proj
      value: 0.06577
    - filter: model.layers.7.mlp.gate_proj
      value: 0.11425
    - filter: model.layers.8.mlp.gate_proj
      value: 0.12968
    - filter: model.layers.9.mlp.gate_proj
      value: 0.19222
    - filter: model.layers.10.mlp.gate_proj
      value: 0.17268
    - filter: model.layers.11.mlp.gate_proj
      value: 0.10512
    - filter: model.layers.12.mlp.gate_proj
      value: 0.02651
    - filter: model.layers.13.mlp.gate_proj
      value: 0.04687
    - filter: model.layers.14.mlp.gate_proj
      value: 0.14716
    - filter: model.layers.15.mlp.gate_proj
      value: 0.03147
    - filter: model.layers.16.mlp.gate_proj
      value: 0.05726
    - filter: model.layers.17.mlp.gate_proj
      value: 0.04511
    - filter: model.layers.18.mlp.gate_proj
      value: 0.25063
    - filter: model.layers.19.mlp.gate_proj
      value: 0.26688
    - filter: model.layers.20.mlp.gate_proj
      value: 0.37716
    - filter: model.layers.21.mlp.gate_proj
      value: 0.3862
    - filter: model.layers.22.mlp.gate_proj
      value: 0.27378
    - filter: model.layers.23.mlp.gate_proj
      value: 0.08641
    - filter: model.layers.24.mlp.gate_proj
      value: 0.37212
    - filter: model.layers.25.mlp.gate_proj
      value: 0.31151
    - filter: model.layers.0.mlp.up_proj
      value: 0.24228
    - filter: model.layers.1.mlp.up_proj
      value: 0.06887
    - filter: model.layers.2.mlp.up_proj
      value: 0.20625
    - filter: model.layers.3.mlp.up_proj
      value: 0.27546
    - filter: model.layers.4.mlp.up_proj
      value: 0.59112
    - filter: model.layers.5.mlp.up_proj
      value: 0.41203
    - filter: model.layers.6.mlp.up_proj
      value: 0.07411
    - filter: model.layers.7.mlp.up_proj
      value: 0.05424
    - filter: model.layers.8.mlp.up_proj
      value: 0.23792
    - filter: model.layers.9.mlp.up_proj
      value: 0.1677
    - filter: model.layers.10.mlp.up_proj
      value: 0.17054
    - filter: model.layers.11.mlp.up_proj
      value: 0.18762
    - filter: model.layers.12.mlp.up_proj
      value: 0.08044
    - filter: model.layers.13.mlp.up_proj
      value: 0.0021
    - filter: model.layers.14.mlp.up_proj
      value: 0.26389
    - filter: model.layers.15.mlp.up_proj
      value: 0.06886
    - filter: model.layers.16.mlp.up_proj
      value: 0.21476
    - filter: model.layers.17.mlp.up_proj
      value: 0.21777
    - filter: model.layers.18.mlp.up_proj
      value: 0.32111
    - filter: model.layers.19.mlp.up_proj
      value: 0.2885
    - filter: model.layers.20.mlp.up_proj
      value: 0.40549
    - filter: model.layers.21.mlp.up_proj
      value: 0.42539
    - filter: model.layers.22.mlp.up_proj
      value: 0.31214
    - filter: model.layers.23.mlp.up_proj
      value: 0.02931
    - filter: model.layers.24.mlp.up_proj
      value: 0.3693
    - filter: model.layers.25.mlp.up_proj
      value: 0.345
    - filter: model.layers.0.mlp.down_proj
      value: 0.06756
    - filter: model.layers.1.mlp.down_proj
      value: 0.03746
    - filter: model.layers.2.mlp.down_proj
      value: 0.09104
    - filter: model.layers.3.mlp.down_proj
      value: 0.06643
    - filter: model.layers.4.mlp.down_proj
      value: 0.05003
    - filter: model.layers.5.mlp.down_proj
      value: 0.0406
    - filter: model.layers.6.mlp.down_proj
      value: 0.01609
    - filter: model.layers.7.mlp.down_proj
      value: 0.09629
    - filter: model.layers.8.mlp.down_proj
      value: 0.08912
    - filter: model.layers.9.mlp.down_proj
      value: 0.12874
    - filter: model.layers.10.mlp.down_proj
      value: 0.04635
    - filter: model.layers.11.mlp.down_proj
      value: 0.0099
    - filter: model.layers.12.mlp.down_proj
      value: 0.03487
    - filter: model.layers.13.mlp.down_proj
      value: 0.04977
    - filter: model.layers.14.mlp.down_proj
      value: 0.00393
    - filter: model.layers.15.mlp.down_proj
      value: 0.00748
    - filter: model.layers.16.mlp.down_proj
      value: 0.06696
    - filter: model.layers.17.mlp.down_proj
      value: 0.02067
    - filter: model.layers.18.mlp.down_proj
      value: 0.10311
    - filter: model.layers.19.mlp.down_proj
      value: 0.009
    - filter: model.layers.20.mlp.down_proj
      value: 0.0215
    - filter: model.layers.21.mlp.down_proj
      value: 0.04196
    - filter: model.layers.22.mlp.down_proj
      value: 0.06326
    - filter: model.layers.23.mlp.down_proj
      value: 0.21737
    - filter: model.layers.24.mlp.down_proj
      value: 0.19338
    - filter: model.layers.25.mlp.down_proj
      value: 0.04905
    weight:
    - value: 1
merge_method: ties
base_model: google/gemma-2-2b
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16
tokenizer_source: union

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "choprahetarth/gemma-instruct-merge-named_correctly"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])