exp-model-parameter-similarity/results/qwen35-2b/analysis/block_sampling.csv

1	role	shape	scale	total_blocks	sampled_blocks
2	model.language_model.embed_tokens.weight	(128, 128)	2d-128	31040	256
3	model.language_model.embed_tokens.weight	(64, 64)	2d-64	124160	256
4	model.language_model.layers.*.linear_attn.conv1d.weight	(1, 1, 4)	3d-axis0-slice	110592	256
5	model.language_model.layers.*.linear_attn.conv1d.weight	(4, 1, 4)	3d-volume	27648	256
6	model.language_model.layers.*.linear_attn.in_proj_qkv.weight	(128, 128)	2d-128	13824	256
7	model.language_model.layers.*.linear_attn.in_proj_qkv.weight	(64, 64)	2d-64	55296	256
8	model.language_model.layers.*.linear_attn.in_proj_z.weight	(128, 128)	2d-128	4608	256
9	model.language_model.layers.*.linear_attn.in_proj_z.weight	(64, 64)	2d-64	18432	256
10	model.language_model.layers.*.linear_attn.out_proj.weight	(128, 128)	2d-128	4608	256
11	model.language_model.layers.*.linear_attn.out_proj.weight	(64, 64)	2d-64	18432	256
12	model.language_model.layers.*.mlp.down_proj.weight	(128, 128)	2d-128	18432	256
13	model.language_model.layers.*.mlp.down_proj.weight	(64, 64)	2d-64	73728	256
14	model.language_model.layers.*.mlp.gate_proj.weight	(128, 128)	2d-128	18432	256
15	model.language_model.layers.*.mlp.gate_proj.weight	(64, 64)	2d-64	73728	256
16	model.language_model.layers.*.mlp.up_proj.weight	(128, 128)	2d-128	18432	256
17	model.language_model.layers.*.mlp.up_proj.weight	(64, 64)	2d-64	73728	256
18	model.language_model.layers.*.self_attn.k_proj.weight	(128, 128)	2d-128	384	256
19	model.language_model.layers.*.self_attn.k_proj.weight	(64, 64)	2d-64	1536	256
20	model.language_model.layers.*.self_attn.o_proj.weight	(128, 128)	2d-128	1536	256
21	model.language_model.layers.*.self_attn.o_proj.weight	(64, 64)	2d-64	6144	256
22	model.language_model.layers.*.self_attn.q_proj.weight	(128, 128)	2d-128	3072	256
23	model.language_model.layers.*.self_attn.q_proj.weight	(64, 64)	2d-64	12288	256
24	model.language_model.layers.*.self_attn.v_proj.weight	(128, 128)	2d-128	384	256
25	model.language_model.layers.*.self_attn.v_proj.weight	(64, 64)	2d-64	1536	256
26	model.visual.layers.*.attn.proj.weight	(128, 128)	2d-128	1536	256
27	model.visual.layers.*.attn.proj.weight	(64, 64)	2d-64	6144	256
28	model.visual.layers.*.attn.qkv.weight	(128, 128)	2d-128	4608	256
29	model.visual.layers.*.attn.qkv.weight	(64, 64)	2d-64	18432	256
30	model.visual.layers.*.mlp.linear_fc1.weight	(128, 128)	2d-128	6144	256
31	model.visual.layers.*.mlp.linear_fc1.weight	(64, 64)	2d-64	24576	256
32	model.visual.layers.*.mlp.linear_fc2.weight	(128, 128)	2d-128	6144	256
33	model.visual.layers.*.mlp.linear_fc2.weight	(64, 64)	2d-64	24576	256
34	model.visual.merger.linear_fc1.weight	(128, 128)	2d-128	1024	256
35	model.visual.merger.linear_fc1.weight	(64, 64)	2d-64	4096	256
36	model.visual.merger.linear_fc2.weight	(128, 128)	2d-128	512	256
37	model.visual.merger.linear_fc2.weight	(64, 64)	2d-64	2048	256
38	model.visual.pos_embed.weight	(128, 128)	2d-128	144	144
39	model.visual.pos_embed.weight	(64, 64)	2d-64	576	256
40	mtp.fc.weight	(128, 128)	2d-128	512	256
41	mtp.fc.weight	(64, 64)	2d-64	2048	256
42	mtp.layers.*.mlp.down_proj.weight	(128, 128)	2d-128	768	256
43	mtp.layers.*.mlp.down_proj.weight	(64, 64)	2d-64	3072	256
44	mtp.layers.*.mlp.gate_proj.weight	(128, 128)	2d-128	768	256
45	mtp.layers.*.mlp.gate_proj.weight	(64, 64)	2d-64	3072	256
46	mtp.layers.*.mlp.up_proj.weight	(128, 128)	2d-128	768	256
47	mtp.layers.*.mlp.up_proj.weight	(64, 64)	2d-64	3072	256
48	mtp.layers.*.self_attn.k_proj.weight	(128, 128)	2d-128	64	64
49	mtp.layers.*.self_attn.k_proj.weight	(64, 64)	2d-64	256	256
50	mtp.layers.*.self_attn.o_proj.weight	(128, 128)	2d-128	256	256
51	mtp.layers.*.self_attn.o_proj.weight	(64, 64)	2d-64	1024	256
52	mtp.layers.*.self_attn.q_proj.weight	(128, 128)	2d-128	512	256
53	mtp.layers.*.self_attn.q_proj.weight	(64, 64)	2d-64	2048	256
54	mtp.layers.*.self_attn.v_proj.weight	(128, 128)	2d-128	64	64
55	mtp.layers.*.self_attn.v_proj.weight	(64, 64)	2d-64	256	256