siddharthmb/2026.TA.l1c_headline_2026-10_h_decoupled_l1_tc8192_decb_l1w0.001_dl1_fp32_tarbb_lb2_sl17811933
4B • Updated
Siddharth M. Bhatia
siddharthmb
·
AI & ML interests
None yet
Recent Activity
updated a model 27 minutes ago
siddharthmb/2026.TA.l1c_headline_2026-10_h_decoupled_l1_tc8192_decb_l1w0.001_dl1_fp32_tarbb_lb2_sl17811933 updated a model 28 minutes ago
siddharthmb/2026.TA.l1c_headline_2026-10_h2_decoupled_l1_tc8192_decb_l1w0.3_dl1_fp32_tarbb_lb2_sl17811934 published a model 28 minutes ago
siddharthmb/2026.TA.l1c_headline_2026-10_h_decoupled_l1_tc8192_decb_l1w0.001_dl1_fp32_tarbb_lb2_sl17811933Organizations
siddharthmb/2026.TA.l1c_headline_2026-10_h2_decoupled_l1_tc8192_decb_l1w0.3_dl1_fp32_tarbb_lb2_sl17811934
4B • Updated
siddharthmb/2026.TA.gemma2_2b_full_500k_l1_0p016_gemma2_2b_full_500k_tc8192_decb_l1w0.016_tarbb_sl15516714
4B • Updated
siddharthmb/2026.TA.gemma2_2b_huge_tc16384_decb_l1w0.0003_norm_sch_tarbb_lb2.0_ln1.0_dr500000_sl15995284
5B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_a_const_tc8192_decb_l1w0.001_fp32_tarbb_lb2_ln1_dr40000_sl17810443
4B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_c_const3e-4_tc8192_decb_l1w0.0003_fp32_tarbb_lb2_ln1_sl17810445
4B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_e_ramp_resample_tc8192_decb_l1w0.001_sch_rs1000_fp32_sl17810509
4B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_b_ramp_tc8192_decb_l1w0.001_sch_fp32_tarbb_lb2_ln1_sl17810444
4B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_f_const_bf16_tc8192_decb_l1w0.001_tarbb_lb2_ln1_dr40000_sl17810536
4B • Updated
siddharthmb/2026.TA.l1c_headline_2026-10_d_ramp3e-4_tc8192_decb_l1w0.0003_sch_fp32_tarbb_lb2_ln1_sl17810446
4B • Updated
siddharthmb/2026.TA.oldlosses_retrain_old_losses_tc8192_decb_l1w0.001_tarbb_lb2.0_ln1_dr20000_sl17794184
4B • Updated • 20
siddharthmb/2026.TA.gemma2_2b_util_probe_tc32768_l1_0p0001_lr4e4_tc32768_decb_l1w0.0001_tarbb_sl15532696
7B • Updated • 14
siddharthmb/2026.TA.gemma2_2b_util_probe_tc16384_l1_0p00025_lr4e4_tc16384_decb_l1w0.00025_tarbb_sl15532697
5B • Updated • 17
siddharthmb/2026.TA.gemma2_2b_util_probe_tc16384_l1_0p0001_lr4e4_tc16384_decb_l1w0.0001_tarbb_sl15532699
5B • Updated • 11
siddharthmb/2026.TA.gemma2_2b_tc8192_decb_l1w0.001_tarbb_lb2.0_ln1_dr20000_lr8e-04_bs4_sl14793860
4B • Updated • 117
siddharthmb/2026.TA.gemma2_2b_tc8192_decb_l1w0.001_tarbb_lb2.0_ln1_dr10000_lr8e-04_bs4_sl14818386
4B • Updated • 14
siddharthmb/2026.TA.gemma2_2b_sparse_probe_tc32768_l1_0p001_tc32768_decb_l1w0.001_tarbb_lb2.0_ln1_sl15529832
7B • Updated • 18
siddharthmb/2026.TA.gemma2_2b_sparse_probe_tc32768_l1_0p0005_tc32768_decb_l1w0.0005_tarbb_lb2.0_sl15529833
7B • Updated • 14
siddharthmb/2026.TA.gemma2_2b_sparse_probe_tc16384_l1_0p001_tc16384_decb_l1w0.001_tarbb_lb2.0_ln1_sl15529831
5B • Updated • 9
siddharthmb/2026.TA.gemma2_2b_sparse_probe_tc16384_l1_0p0005_tc16384_decb_l1w0.0005_tarbb_lb2.0_sl15529830
5B • Updated • 11
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p128_tc8192_decb_l1w0.128_tarbb_lb2.0_ln1_dr10000_sl15516048
4B • Updated • 15
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p064_tc8192_decb_l1w0.064_tarbb_lb2.0_ln1_dr10000_sl15516047
4B • Updated • 23
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p048_tc8192_decb_l1w0.048_tarbb_lb2.0_ln1_dr10000_sl15516046
4B • Updated • 12
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p032_tc8192_decb_l1w0.032_tarbb_lb2.0_ln1_dr10000_sl15516045
4B • Updated • 25
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p016_tc8192_decb_l1w0.016_tarbb_lb2.0_ln1_dr10000_sl15512630
4B • Updated • 15
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p008_tc8192_decb_l1w0.008_tarbb_lb2.0_ln1_dr10000_sl15512629
4B • Updated • 19 • 1
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p004_tc8192_decb_l1w0.004_tarbb_lb2.0_ln1_dr10000_sl15512628
4B • Updated • 19
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p002_tc8192_decb_l1w0.002_tarbb_lb2.0_ln1_dr10000_sl15512627
4B • Updated • 16
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p001_tc8192_decb_l1w0.001_tarbb_lb2.0_ln1_dr10000_sl15529809
4B • Updated • 22
siddharthmb/2026.TA.gemma2_2b_sparse_probe_l1_0p0015_tc8192_decb_l1w0.0015_tarbb_lb2.0_ln1_sl15529808
4B • Updated • 21