Views
No views yet
meta-llama/Llama-3.2-3BIsoCTS./merged_models/Llama-3.2-3B_merged/IsoCTS_scaling_coef_1_common_space_fraction_0.8_validate_gen_flag_True_validate_scaling_start_0.0_validate_scaling_stop_1.0_validate_scaling_step_0.21{
2 "local_rank": 0,
3 "scaling_coef": 1,
4 "K": 0.2,
5 "merge_func": "sum",
6 "p": 0.8,
7 "sigmoid_bias": 3,
8 "sparsity": 0.1,
9 "lr": 100000000.0,
10 "n_epochs": 1,
11 "l1_strength": 0.0,
12 "dataless": false,
13 "k": 2,
14 "lamda": [
15 0.5
16 ],
17 "lamda_tuning": false,
18 "task_names": null,
19 "reduction": null,
20 "fisher_only": false,
21 "merge_only": false,
22 "save_group": null,
23 "model_coeff_value": 0.3,
24 "keep_checkpoints": false,
25 "beta_coef": 1,
26 "n": 64,
27 "m": 256,
28 "pruned_subdir": "pruned_models",
29 "save_pruned_models": true,
30 "warn_min_overlap_fallback": true,
31 "common_space_fraction": 0.8,
32 "validate_gen_flag": true,
33 "validate_scaling_start": 0.0,
34 "validate_scaling_stop": 1.0,
35 "validate_scaling_step": 0.2,
36 "base_model": "meta-llama/Llama-3.2-3B",
37 "algo": "IsoCTS",
38 "save_path": "./merged_models/"
39}1{
2 "scaling_coef": 0.6,
3 "common_space_fraction": 0.8,
4 "validate_gen_flag": true,
5 "validate_scaling_start": 0.0,
6 "validate_scaling_stop": 1.0,
7 "validate_scaling_step": 0.2
8}