{ "lm_head.weight": { "numel": 1271398400, "differing": 392702, "frac": 0.00031, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.389464513034682e-10, "official_absmax": 0.341796875 }, "model.language_model.layers.0.input_layernorm.weight": { "numel": 5120, "differing": 2732, "frac": 0.53359, "max_abs": 0.00048828125, "rel_l1": 0.002601978136226535, "official_absmax": 0.1982421875 }, "model.language_model.layers.0.linear_attn.conv1d.weight": { "numel": 40960, "differing": 42, "frac": 0.00103, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.3925084418886513e-09, "official_absmax": 0.51171875 }, "model.language_model.layers.0.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 19.25 }, "model.language_model.layers.0.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9296875 }, "model.language_model.layers.0.post_attention_layernorm.weight": { "numel": 5120, "differing": 5, "frac": 0.00098, "max_abs": 0.000244140625, "rel_l1": 1.0286789802194107e-06, "official_absmax": 0.99609375 }, "model.language_model.layers.1.input_layernorm.weight": { "numel": 5120, "differing": 3203, "frac": 0.62559, "max_abs": 0.00048828125, "rel_l1": 0.0029295964632183313, "official_absmax": 0.671875 }, "model.language_model.layers.1.linear_attn.conv1d.weight": { "numel": 40960, "differing": 32, "frac": 0.00078, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1900677110432412e-09, "official_absmax": 0.49609375 }, "model.language_model.layers.1.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.9375 }, "model.language_model.layers.1.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.95703125 }, "model.language_model.layers.1.post_attention_layernorm.weight": { "numel": 5120, "differing": 1, "frac": 0.0002, "max_abs": 0.00042724609375, "rel_l1": 5.271208465273958e-07, "official_absmax": 0.99609375 }, "model.language_model.layers.10.input_layernorm.weight": { "numel": 5120, "differing": 3489, "frac": 0.68145, "max_abs": 0.00048828125, "rel_l1": 0.00371382012963295, "official_absmax": 0.55859375 }, "model.language_model.layers.10.linear_attn.conv1d.weight": { "numel": 40960, "differing": 27, "frac": 0.00066, "max_abs": 2.9802322387695312e-08, "rel_l1": 8.8305518453069e-10, "official_absmax": 0.8828125 }, "model.language_model.layers.10.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.5 }, "model.language_model.layers.10.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9765625 }, "model.language_model.layers.10.post_attention_layernorm.weight": { "numel": 5120, "differing": 1826, "frac": 0.35664, "max_abs": 0.00048828125, "rel_l1": 0.0008708190871402621, "official_absmax": 0.99609375 }, "model.language_model.layers.11.input_layernorm.weight": { "numel": 5120, "differing": 2966, "frac": 0.5793, "max_abs": 0.00048828125, "rel_l1": 0.0020617882255464792, "official_absmax": 1.0078125 }, "model.language_model.layers.11.post_attention_layernorm.weight": { "numel": 5120, "differing": 2990, "frac": 0.58398, "max_abs": 0.00048828125, "rel_l1": 0.0020256787538528442, "official_absmax": 0.92578125 }, "model.language_model.layers.11.self_attn.k_norm.weight": { "numel": 256, "differing": 9, "frac": 0.03516, "max_abs": 0.00048828125, "rel_l1": 3.0420042094192468e-05, "official_absmax": 0.7109375 }, "model.language_model.layers.11.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.000244140625, "rel_l1": 2.6353864086559042e-06, "official_absmax": 0.52734375 }, "model.language_model.layers.12.input_layernorm.weight": { "numel": 5120, "differing": 3432, "frac": 0.67031, "max_abs": 0.00048828125, "rel_l1": 0.00367084052413702, "official_absmax": 0.470703125 }, "model.language_model.layers.12.linear_attn.conv1d.weight": { "numel": 40960, "differing": 22, "frac": 0.00054, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.019635345744234e-10, "official_absmax": 0.5234375 }, "model.language_model.layers.12.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.59375 }, "model.language_model.layers.12.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0234375 }, "model.language_model.layers.12.post_attention_layernorm.weight": { "numel": 5120, "differing": 2759, "frac": 0.53887, "max_abs": 0.00048828125, "rel_l1": 0.0018413743237033486, "official_absmax": 0.93359375 }, "model.language_model.layers.13.input_layernorm.weight": { "numel": 5120, "differing": 3306, "frac": 0.6457, "max_abs": 0.00048828125, "rel_l1": 0.003474080003798008, "official_absmax": 0.4609375 }, "model.language_model.layers.13.linear_attn.conv1d.weight": { "numel": 40960, "differing": 26, "frac": 0.00063, "max_abs": 2.9802322387695312e-08, "rel_l1": 8.241581306300816e-10, "official_absmax": 0.490234375 }, "model.language_model.layers.13.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.21875 }, "model.language_model.layers.13.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0390625 }, "model.language_model.layers.13.post_attention_layernorm.weight": { "numel": 5120, "differing": 2361, "frac": 0.46113, "max_abs": 0.00048828125, "rel_l1": 0.001400968641974032, "official_absmax": 0.9375 }, "model.language_model.layers.14.input_layernorm.weight": { "numel": 5120, "differing": 3623, "frac": 0.70762, "max_abs": 0.00048828125, "rel_l1": 0.004010478034615517, "official_absmax": 0.466796875 }, "model.language_model.layers.14.linear_attn.conv1d.weight": { "numel": 40960, "differing": 24, "frac": 0.00059, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.84980572263072e-10, "official_absmax": 0.609375 }, "model.language_model.layers.14.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.8125 }, "model.language_model.layers.14.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.015625 }, "model.language_model.layers.14.post_attention_layernorm.weight": { "numel": 5120, "differing": 1842, "frac": 0.35977, "max_abs": 0.00048828125, "rel_l1": 0.0009047642815858126, "official_absmax": 0.95703125 }, "model.language_model.layers.15.input_layernorm.weight": { "numel": 5120, "differing": 3194, "frac": 0.62383, "max_abs": 0.00048828125, "rel_l1": 0.002849259879440069, "official_absmax": 0.8984375 }, "model.language_model.layers.15.post_attention_layernorm.weight": { "numel": 5120, "differing": 2566, "frac": 0.50117, "max_abs": 0.00048828125, "rel_l1": 0.001617958303540945, "official_absmax": 0.92578125 }, "model.language_model.layers.15.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 8.799419447313994e-06, "official_absmax": 0.84765625 }, "model.language_model.layers.15.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00048828125, "rel_l1": 4.957243618264329e-06, "official_absmax": 0.8046875 }, "model.language_model.layers.16.input_layernorm.weight": { "numel": 5120, "differing": 3667, "frac": 0.71621, "max_abs": 0.00048828125, "rel_l1": 0.004291723947972059, "official_absmax": 0.458984375 }, "model.language_model.layers.16.linear_attn.conv1d.weight": { "numel": 40960, "differing": 27, "frac": 0.00066, "max_abs": 2.9802322387695312e-08, "rel_l1": 9.098483633174226e-10, "official_absmax": 0.6640625 }, "model.language_model.layers.16.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.90625 }, "model.language_model.layers.16.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0 }, "model.language_model.layers.16.post_attention_layernorm.weight": { "numel": 5120, "differing": 2278, "frac": 0.44492, "max_abs": 0.00048828125, "rel_l1": 0.0014892446342855692, "official_absmax": 0.9140625 }, "model.language_model.layers.17.input_layernorm.weight": { "numel": 5120, "differing": 3657, "frac": 0.71426, "max_abs": 0.00048828125, "rel_l1": 0.004427246283739805, "official_absmax": 0.435546875 }, "model.language_model.layers.17.linear_attn.conv1d.weight": { "numel": 40960, "differing": 28, "frac": 0.00068, "max_abs": 2.9802322387695312e-08, "rel_l1": 8.588754152327738e-10, "official_absmax": 0.54296875 }, "model.language_model.layers.17.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.53125 }, "model.language_model.layers.17.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9921875 }, "model.language_model.layers.17.post_attention_layernorm.weight": { "numel": 5120, "differing": 2520, "frac": 0.49219, "max_abs": 0.00048828125, "rel_l1": 0.0018250048160552979, "official_absmax": 0.90234375 }, "model.language_model.layers.18.input_layernorm.weight": { "numel": 5120, "differing": 3867, "frac": 0.75527, "max_abs": 0.00048828125, "rel_l1": 0.005019516684114933, "official_absmax": 0.4296875 }, "model.language_model.layers.18.linear_attn.conv1d.weight": { "numel": 40960, "differing": 23, "frac": 0.00056, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.379778927367852e-10, "official_absmax": 0.498046875 }, "model.language_model.layers.18.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.53125 }, "model.language_model.layers.18.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9921875 }, "model.language_model.layers.18.post_attention_layernorm.weight": { "numel": 5120, "differing": 2215, "frac": 0.43262, "max_abs": 0.00048828125, "rel_l1": 0.0017191976075991988, "official_absmax": 0.84375 }, "model.language_model.layers.19.input_layernorm.weight": { "numel": 5120, "differing": 2647, "frac": 0.51699, "max_abs": 0.00048828125, "rel_l1": 0.0020424339454621077, "official_absmax": 0.56640625 }, "model.language_model.layers.19.post_attention_layernorm.weight": { "numel": 5120, "differing": 3397, "frac": 0.66348, "max_abs": 0.00048828125, "rel_l1": 0.0027294810861349106, "official_absmax": 0.828125 }, "model.language_model.layers.19.self_attn.k_norm.weight": { "numel": 256, "differing": 9, "frac": 0.03516, "max_abs": 0.00048828125, "rel_l1": 2.6473811885807663e-05, "official_absmax": 0.76171875 }, "model.language_model.layers.19.self_attn.q_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 9.371017767989542e-06, "official_absmax": 0.71484375 }, "model.language_model.layers.2.input_layernorm.weight": { "numel": 5120, "differing": 3754, "frac": 0.7332, "max_abs": 0.00048828125, "rel_l1": 0.0045973993837833405, "official_absmax": 0.640625 }, "model.language_model.layers.2.linear_attn.conv1d.weight": { "numel": 40960, "differing": 28, "frac": 0.00068, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.0271316019938581e-09, "official_absmax": 0.388671875 }, "model.language_model.layers.2.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.65625 }, "model.language_model.layers.2.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9921875 }, "model.language_model.layers.2.post_attention_layernorm.weight": { "numel": 5120, "differing": 1, "frac": 0.0002, "max_abs": 0.0001220703125, "rel_l1": 1.582051396553652e-07, "official_absmax": 0.99609375 }, "model.language_model.layers.20.input_layernorm.weight": { "numel": 5120, "differing": 2308, "frac": 0.45078, "max_abs": 0.00048828125, "rel_l1": 0.0020661414600908756, "official_absmax": 0.515625 }, "model.language_model.layers.20.linear_attn.conv1d.weight": { "numel": 40960, "differing": 23, "frac": 0.00056, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.48273557163148e-10, "official_absmax": 0.66015625 }, "model.language_model.layers.20.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.5625 }, "model.language_model.layers.20.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0078125 }, "model.language_model.layers.20.post_attention_layernorm.weight": { "numel": 5120, "differing": 3378, "frac": 0.65977, "max_abs": 0.00048828125, "rel_l1": 0.00259850500151515, "official_absmax": 0.84765625 }, "model.language_model.layers.21.input_layernorm.weight": { "numel": 5120, "differing": 777, "frac": 0.15176, "max_abs": 0.00048828125, "rel_l1": 0.0004565082781482488, "official_absmax": 0.427734375 }, "model.language_model.layers.21.linear_attn.conv1d.weight": { "numel": 40960, "differing": 16, "frac": 0.00039, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.3682316097637397e-10, "official_absmax": 0.62109375 }, "model.language_model.layers.21.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.84375 }, "model.language_model.layers.21.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0078125 }, "model.language_model.layers.21.post_attention_layernorm.weight": { "numel": 5120, "differing": 3244, "frac": 0.63359, "max_abs": 0.00048828125, "rel_l1": 0.0022902546916157007, "official_absmax": 0.859375 }, "model.language_model.layers.22.input_layernorm.weight": { "numel": 5120, "differing": 1952, "frac": 0.38125, "max_abs": 0.00048828125, "rel_l1": 0.0014872492756694555, "official_absmax": 0.5234375 }, "model.language_model.layers.22.linear_attn.conv1d.weight": { "numel": 40960, "differing": 32, "frac": 0.00078, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.013276684780351e-09, "official_absmax": 0.62109375 }, "model.language_model.layers.22.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.6875 }, "model.language_model.layers.22.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0390625 }, "model.language_model.layers.22.post_attention_layernorm.weight": { "numel": 5120, "differing": 2954, "frac": 0.57695, "max_abs": 0.00048828125, "rel_l1": 0.0021364109124988317, "official_absmax": 0.77734375 }, "model.language_model.layers.23.input_layernorm.weight": { "numel": 5120, "differing": 3287, "frac": 0.64199, "max_abs": 0.00048828125, "rel_l1": 0.003475325647741556, "official_absmax": 0.6171875 }, "model.language_model.layers.23.post_attention_layernorm.weight": { "numel": 5120, "differing": 2691, "frac": 0.52559, "max_abs": 0.00048828125, "rel_l1": 0.0019086959073320031, "official_absmax": 0.78125 }, "model.language_model.layers.23.self_attn.k_norm.weight": { "numel": 256, "differing": 4, "frac": 0.01562, "max_abs": 0.00048828125, "rel_l1": 8.17724230728345e-06, "official_absmax": 1.203125 }, "model.language_model.layers.23.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00048828125, "rel_l1": 4.1860444071062375e-06, "official_absmax": 0.87109375 }, "model.language_model.layers.24.input_layernorm.weight": { "numel": 5120, "differing": 3971, "frac": 0.77559, "max_abs": 0.00048828125, "rel_l1": 0.005521549843251705, "official_absmax": 0.55078125 }, "model.language_model.layers.24.linear_attn.conv1d.weight": { "numel": 40960, "differing": 29, "frac": 0.00071, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.881857944092019e-10, "official_absmax": 0.5546875 }, "model.language_model.layers.24.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.53125 }, "model.language_model.layers.24.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0703125 }, "model.language_model.layers.24.post_attention_layernorm.weight": { "numel": 5120, "differing": 2264, "frac": 0.44219, "max_abs": 0.00048828125, "rel_l1": 0.0014577610418200493, "official_absmax": 0.71875 }, "model.language_model.layers.25.input_layernorm.weight": { "numel": 5120, "differing": 3523, "frac": 0.68809, "max_abs": 0.00048828125, "rel_l1": 0.0036650600377470255, "official_absmax": 0.640625 }, "model.language_model.layers.25.linear_attn.conv1d.weight": { "numel": 40960, "differing": 39, "frac": 0.00095, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.2213549061002027e-09, "official_absmax": 0.447265625 }, "model.language_model.layers.25.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.8125 }, "model.language_model.layers.25.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0859375 }, "model.language_model.layers.25.post_attention_layernorm.weight": { "numel": 5120, "differing": 1938, "frac": 0.37852, "max_abs": 0.00048828125, "rel_l1": 0.0011293068528175354, "official_absmax": 0.71875 }, "model.language_model.layers.26.input_layernorm.weight": { "numel": 5120, "differing": 3087, "frac": 0.60293, "max_abs": 0.00048828125, "rel_l1": 0.002827809890732169, "official_absmax": 0.6484375 }, "model.language_model.layers.26.linear_attn.conv1d.weight": { "numel": 40960, "differing": 36, "frac": 0.00088, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.0085779988955323e-09, "official_absmax": 0.6171875 }, "model.language_model.layers.26.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.71875 }, "model.language_model.layers.26.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.203125 }, "model.language_model.layers.26.post_attention_layernorm.weight": { "numel": 5120, "differing": 2033, "frac": 0.39707, "max_abs": 0.00048828125, "rel_l1": 0.0011633371468633413, "official_absmax": 0.72265625 }, "model.language_model.layers.27.input_layernorm.weight": { "numel": 5120, "differing": 2225, "frac": 0.43457, "max_abs": 0.00048828125, "rel_l1": 0.0016584779368713498, "official_absmax": 0.6328125 }, "model.language_model.layers.27.post_attention_layernorm.weight": { "numel": 5120, "differing": 1617, "frac": 0.31582, "max_abs": 0.00048828125, "rel_l1": 0.0008193491958081722, "official_absmax": 0.72265625 }, "model.language_model.layers.27.self_attn.k_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 4.946499302604934e-06, "official_absmax": 1.5859375 }, "model.language_model.layers.27.self_attn.q_norm.weight": { "numel": 256, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.3671875 }, "model.language_model.layers.28.input_layernorm.weight": { "numel": 5120, "differing": 2055, "frac": 0.40137, "max_abs": 0.00048828125, "rel_l1": 0.0011943163117393851, "official_absmax": 0.6953125 }, "model.language_model.layers.28.linear_attn.conv1d.weight": { "numel": 40960, "differing": 33, "frac": 0.00081, "max_abs": 2.9802322387695312e-08, "rel_l1": 9.210072704490813e-10, "official_absmax": 0.7734375 }, "model.language_model.layers.28.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.5 }, "model.language_model.layers.28.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.1640625 }, "model.language_model.layers.28.post_attention_layernorm.weight": { "numel": 5120, "differing": 1595, "frac": 0.31152, "max_abs": 0.00048828125, "rel_l1": 0.0007244602311402559, "official_absmax": 0.7265625 }, "model.language_model.layers.29.input_layernorm.weight": { "numel": 5120, "differing": 1750, "frac": 0.3418, "max_abs": 0.00048828125, "rel_l1": 0.000852582510560751, "official_absmax": 0.70703125 }, "model.language_model.layers.29.linear_attn.conv1d.weight": { "numel": 40960, "differing": 29, "frac": 0.00071, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.798958145954771e-10, "official_absmax": 0.56640625 }, "model.language_model.layers.29.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.53125 }, "model.language_model.layers.29.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.28125 }, "model.language_model.layers.29.post_attention_layernorm.weight": { "numel": 5120, "differing": 1475, "frac": 0.28809, "max_abs": 0.00048828125, "rel_l1": 0.0006210694555193186, "official_absmax": 0.734375 }, "model.language_model.layers.3.input_layernorm.weight": { "numel": 5120, "differing": 6, "frac": 0.00117, "max_abs": 0.00048828125, "rel_l1": 1.8172318050346803e-06, "official_absmax": 0.890625 }, "model.language_model.layers.3.post_attention_layernorm.weight": { "numel": 5120, "differing": 142, "frac": 0.02773, "max_abs": 0.00048828125, "rel_l1": 6.311525066848844e-05, "official_absmax": 0.99609375 }, "model.language_model.layers.3.self_attn.k_norm.weight": { "numel": 256, "differing": 7, "frac": 0.02734, "max_abs": 0.00048828125, "rel_l1": 3.507896690280177e-05, "official_absmax": 0.75390625 }, "model.language_model.layers.3.self_attn.k_proj.weight": { "numel": 5242880, "differing": 1434, "frac": 0.00027, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.860622460078901e-10, "official_absmax": 0.2099609375 }, "model.language_model.layers.3.self_attn.q_norm.weight": { "numel": 256, "differing": 11, "frac": 0.04297, "max_abs": 0.00048828125, "rel_l1": 6.53205206617713e-05, "official_absmax": 0.48046875 }, "model.language_model.layers.3.self_attn.q_proj.weight": { "numel": 62914560, "differing": 15806, "frac": 0.00025, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.1718140053603747e-10, "official_absmax": 0.3203125 }, "model.language_model.layers.3.self_attn.v_proj.weight": { "numel": 5242880, "differing": 1348, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.3694337037436526e-10, "official_absmax": 0.2265625 }, "model.language_model.layers.30.input_layernorm.weight": { "numel": 5120, "differing": 1621, "frac": 0.3166, "max_abs": 0.00048828125, "rel_l1": 0.0007295055547729135, "official_absmax": 0.71484375 }, "model.language_model.layers.30.linear_attn.conv1d.weight": { "numel": 40960, "differing": 43, "frac": 0.00105, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1979208736079272e-09, "official_absmax": 0.60546875 }, "model.language_model.layers.30.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.0 }, "model.language_model.layers.30.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.265625 }, "model.language_model.layers.30.post_attention_layernorm.weight": { "numel": 5120, "differing": 1536, "frac": 0.3, "max_abs": 0.00048828125, "rel_l1": 0.0006242109811864793, "official_absmax": 0.7890625 }, "model.language_model.layers.31.input_layernorm.weight": { "numel": 5120, "differing": 1727, "frac": 0.3373, "max_abs": 0.00048828125, "rel_l1": 0.0008732745773158967, "official_absmax": 0.78515625 }, "model.language_model.layers.31.post_attention_layernorm.weight": { "numel": 5120, "differing": 1454, "frac": 0.28398, "max_abs": 0.00048828125, "rel_l1": 0.0006044442416168749, "official_absmax": 0.79296875 }, "model.language_model.layers.31.self_attn.k_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 4.975330739398487e-06, "official_absmax": 1.890625 }, "model.language_model.layers.31.self_attn.k_proj.weight": { "numel": 5242880, "differing": 1497, "frac": 0.00029, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.347804976629789e-10, "official_absmax": 0.326171875 }, "model.language_model.layers.31.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.000244140625, "rel_l1": 1.654935886108433e-06, "official_absmax": 1.4765625 }, "model.language_model.layers.31.self_attn.q_proj.weight": { "numel": 62914560, "differing": 18146, "frac": 0.00029, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.483226650504491e-10, "official_absmax": 0.322265625 }, "model.language_model.layers.31.self_attn.v_proj.weight": { "numel": 5242880, "differing": 1381, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.24733553883172e-10, "official_absmax": 0.419921875 }, "model.language_model.layers.32.input_layernorm.weight": { "numel": 5120, "differing": 1568, "frac": 0.30625, "max_abs": 0.00048828125, "rel_l1": 0.0006935849087312818, "official_absmax": 0.72265625 }, "model.language_model.layers.32.linear_attn.conv1d.weight": { "numel": 40960, "differing": 62, "frac": 0.00151, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.7018081388542328e-09, "official_absmax": 0.921875 }, "model.language_model.layers.32.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.59375 }, "model.language_model.layers.32.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.203125 }, "model.language_model.layers.32.post_attention_layernorm.weight": { "numel": 5120, "differing": 1587, "frac": 0.30996, "max_abs": 0.00048828125, "rel_l1": 0.0006950939423404634, "official_absmax": 0.75 }, "model.language_model.layers.33.input_layernorm.weight": { "numel": 5120, "differing": 1718, "frac": 0.33555, "max_abs": 0.00048828125, "rel_l1": 0.0008543466101400554, "official_absmax": 0.68359375 }, "model.language_model.layers.33.linear_attn.conv1d.weight": { "numel": 40960, "differing": 53, "frac": 0.00129, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.4404426540437498e-09, "official_absmax": 0.80859375 }, "model.language_model.layers.33.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.5625 }, "model.language_model.layers.33.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.1640625 }, "model.language_model.layers.33.post_attention_layernorm.weight": { "numel": 5120, "differing": 1816, "frac": 0.35469, "max_abs": 0.00048828125, "rel_l1": 0.0009235324687324464, "official_absmax": 0.65625 }, "model.language_model.layers.34.input_layernorm.weight": { "numel": 5120, "differing": 1916, "frac": 0.37422, "max_abs": 0.00048828125, "rel_l1": 0.0010434159776195884, "official_absmax": 0.66015625 }, "model.language_model.layers.34.linear_attn.conv1d.weight": { "numel": 40960, "differing": 40, "frac": 0.00098, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.06412656464272e-09, "official_absmax": 0.6953125 }, "model.language_model.layers.34.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.5 }, "model.language_model.layers.34.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0703125 }, "model.language_model.layers.34.post_attention_layernorm.weight": { "numel": 5120, "differing": 2108, "frac": 0.41172, "max_abs": 0.00048828125, "rel_l1": 0.001300742500461638, "official_absmax": 0.61328125 }, "model.language_model.layers.35.input_layernorm.weight": { "numel": 5120, "differing": 2670, "frac": 0.52148, "max_abs": 0.00048828125, "rel_l1": 0.0020165250170975924, "official_absmax": 0.765625 }, "model.language_model.layers.35.post_attention_layernorm.weight": { "numel": 5120, "differing": 2447, "frac": 0.47793, "max_abs": 0.00048828125, "rel_l1": 0.0016649194294586778, "official_absmax": 0.65625 }, "model.language_model.layers.35.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 8.039210115384776e-06, "official_absmax": 1.6953125 }, "model.language_model.layers.35.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00048828125, "rel_l1": 3.0869819056533743e-06, "official_absmax": 1.15625 }, "model.language_model.layers.36.input_layernorm.weight": { "numel": 5120, "differing": 1848, "frac": 0.36094, "max_abs": 0.00048828125, "rel_l1": 0.001255508977919817, "official_absmax": 0.59375 }, "model.language_model.layers.36.linear_attn.conv1d.weight": { "numel": 40960, "differing": 37, "frac": 0.0009, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.258652012787081e-10, "official_absmax": 0.66796875 }, "model.language_model.layers.36.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.53125 }, "model.language_model.layers.36.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.046875 }, "model.language_model.layers.36.post_attention_layernorm.weight": { "numel": 5120, "differing": 2687, "frac": 0.5248, "max_abs": 0.00048828125, "rel_l1": 0.0018290275475010276, "official_absmax": 0.7578125 }, "model.language_model.layers.37.input_layernorm.weight": { "numel": 5120, "differing": 1751, "frac": 0.34199, "max_abs": 0.00048828125, "rel_l1": 0.0012085885973647237, "official_absmax": 0.5390625 }, "model.language_model.layers.37.linear_attn.conv1d.weight": { "numel": 40960, "differing": 24, "frac": 0.00059, "max_abs": 2.9802322387695312e-08, "rel_l1": 3.7921057960410565e-10, "official_absmax": 0.64453125 }, "model.language_model.layers.37.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.84375 }, "model.language_model.layers.37.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0 }, "model.language_model.layers.37.post_attention_layernorm.weight": { "numel": 5120, "differing": 2767, "frac": 0.54043, "max_abs": 0.00048828125, "rel_l1": 0.0018927785567939281, "official_absmax": 0.78125 }, "model.language_model.layers.38.input_layernorm.weight": { "numel": 5120, "differing": 1683, "frac": 0.32871, "max_abs": 0.00048828125, "rel_l1": 0.0011095908703282475, "official_absmax": 0.53515625 }, "model.language_model.layers.38.linear_attn.conv1d.weight": { "numel": 40960, "differing": 41, "frac": 0.001, "max_abs": 2.9802322387695312e-08, "rel_l1": 9.831860880993304e-10, "official_absmax": 0.65625 }, "model.language_model.layers.38.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.25 }, "model.language_model.layers.38.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.15625 }, "model.language_model.layers.38.post_attention_layernorm.weight": { "numel": 5120, "differing": 2387, "frac": 0.46621, "max_abs": 0.00048828125, "rel_l1": 0.0016986682312563062, "official_absmax": 0.69140625 }, "model.language_model.layers.39.input_layernorm.weight": { "numel": 5120, "differing": 3689, "frac": 0.72051, "max_abs": 0.00048828125, "rel_l1": 0.004039749503135681, "official_absmax": 0.6875 }, "model.language_model.layers.39.post_attention_layernorm.weight": { "numel": 5120, "differing": 2405, "frac": 0.46973, "max_abs": 0.00048828125, "rel_l1": 0.0016806380590423942, "official_absmax": 0.75390625 }, "model.language_model.layers.39.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 5.323165623849491e-06, "official_absmax": 1.84375 }, "model.language_model.layers.39.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00048828125, "rel_l1": 3.0698670343554113e-06, "official_absmax": 1.3125 }, "model.language_model.layers.4.input_layernorm.weight": { "numel": 5120, "differing": 2030, "frac": 0.39648, "max_abs": 0.00048828125, "rel_l1": 0.0017946554580703378, "official_absmax": 0.703125 }, "model.language_model.layers.4.linear_attn.conv1d.weight": { "numel": 40960, "differing": 34, "frac": 0.00083, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.220471168572601e-09, "official_absmax": 0.93359375 }, "model.language_model.layers.4.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.125 }, "model.language_model.layers.4.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.03125 }, "model.language_model.layers.4.post_attention_layernorm.weight": { "numel": 5120, "differing": 1175, "frac": 0.22949, "max_abs": 0.00048828125, "rel_l1": 0.0006154662114568055, "official_absmax": 0.99609375 }, "model.language_model.layers.40.input_layernorm.weight": { "numel": 5120, "differing": 3336, "frac": 0.65156, "max_abs": 0.00048828125, "rel_l1": 0.003574900794774294, "official_absmax": 0.494140625 }, "model.language_model.layers.40.linear_attn.conv1d.weight": { "numel": 40960, "differing": 29, "frac": 0.00071, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.593898760343109e-10, "official_absmax": 0.65234375 }, "model.language_model.layers.40.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.375 }, "model.language_model.layers.40.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.203125 }, "model.language_model.layers.40.post_attention_layernorm.weight": { "numel": 5120, "differing": 1904, "frac": 0.37188, "max_abs": 0.00048828125, "rel_l1": 0.0011350640561431646, "official_absmax": 0.72265625 }, "model.language_model.layers.41.input_layernorm.weight": { "numel": 5120, "differing": 2728, "frac": 0.53281, "max_abs": 0.00048828125, "rel_l1": 0.002249465323984623, "official_absmax": 0.60546875 }, "model.language_model.layers.41.linear_attn.conv1d.weight": { "numel": 40960, "differing": 31, "frac": 0.00076, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.671109303331036e-10, "official_absmax": 0.62109375 }, "model.language_model.layers.41.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.5 }, "model.language_model.layers.41.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.3046875 }, "model.language_model.layers.41.post_attention_layernorm.weight": { "numel": 5120, "differing": 1704, "frac": 0.33281, "max_abs": 0.00048828125, "rel_l1": 0.0009236435871571302, "official_absmax": 0.71875 }, "model.language_model.layers.42.input_layernorm.weight": { "numel": 5120, "differing": 2895, "frac": 0.56543, "max_abs": 0.00048828125, "rel_l1": 0.0024464051239192486, "official_absmax": 0.64453125 }, "model.language_model.layers.42.linear_attn.conv1d.weight": { "numel": 40960, "differing": 39, "frac": 0.00095, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1058369775440724e-09, "official_absmax": 0.62890625 }, "model.language_model.layers.42.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.59375 }, "model.language_model.layers.42.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.34375 }, "model.language_model.layers.42.post_attention_layernorm.weight": { "numel": 5120, "differing": 1861, "frac": 0.36348, "max_abs": 0.00048828125, "rel_l1": 0.0010608999291434884, "official_absmax": 0.70703125 }, "model.language_model.layers.43.input_layernorm.weight": { "numel": 5120, "differing": 2488, "frac": 0.48594, "max_abs": 0.00048828125, "rel_l1": 0.001944551826454699, "official_absmax": 0.59765625 }, "model.language_model.layers.43.post_attention_layernorm.weight": { "numel": 5120, "differing": 1525, "frac": 0.29785, "max_abs": 0.00048828125, "rel_l1": 0.0007591333123855293, "official_absmax": 0.65625 }, "model.language_model.layers.43.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 7.409481895592762e-06, "official_absmax": 2.140625 }, "model.language_model.layers.43.self_attn.q_norm.weight": { "numel": 256, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.4140625 }, "model.language_model.layers.44.input_layernorm.weight": { "numel": 5120, "differing": 2314, "frac": 0.45195, "max_abs": 0.00048828125, "rel_l1": 0.0015839632833376527, "official_absmax": 0.64453125 }, "model.language_model.layers.44.linear_attn.conv1d.weight": { "numel": 40960, "differing": 26, "frac": 0.00063, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.355447834671679e-10, "official_absmax": 0.7109375 }, "model.language_model.layers.44.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.34375 }, "model.language_model.layers.44.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.2734375 }, "model.language_model.layers.44.post_attention_layernorm.weight": { "numel": 5120, "differing": 1586, "frac": 0.30977, "max_abs": 0.00048828125, "rel_l1": 0.0007749698706902564, "official_absmax": 0.71484375 }, "model.language_model.layers.45.input_layernorm.weight": { "numel": 5120, "differing": 1816, "frac": 0.35469, "max_abs": 0.00048828125, "rel_l1": 0.000995546579360962, "official_absmax": 0.66015625 }, "model.language_model.layers.45.linear_attn.conv1d.weight": { "numel": 40960, "differing": 37, "frac": 0.0009, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.0408017780960677e-09, "official_absmax": 0.734375 }, "model.language_model.layers.45.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.1875 }, "model.language_model.layers.45.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.421875 }, "model.language_model.layers.45.post_attention_layernorm.weight": { "numel": 5120, "differing": 1478, "frac": 0.28867, "max_abs": 0.00048828125, "rel_l1": 0.0006978310993872583, "official_absmax": 0.69140625 }, "model.language_model.layers.46.input_layernorm.weight": { "numel": 5120, "differing": 1913, "frac": 0.37363, "max_abs": 0.00048828125, "rel_l1": 0.0010642192792147398, "official_absmax": 0.66796875 }, "model.language_model.layers.46.linear_attn.conv1d.weight": { "numel": 40960, "differing": 43, "frac": 0.00105, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.0325140742395433e-09, "official_absmax": 0.6796875 }, "model.language_model.layers.46.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.0 }, "model.language_model.layers.46.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.3828125 }, "model.language_model.layers.46.post_attention_layernorm.weight": { "numel": 5120, "differing": 1755, "frac": 0.34277, "max_abs": 0.00048828125, "rel_l1": 0.0008803596720099449, "official_absmax": 0.74609375 }, "model.language_model.layers.47.input_layernorm.weight": { "numel": 5120, "differing": 2505, "frac": 0.48926, "max_abs": 0.00048828125, "rel_l1": 0.0017156848916783929, "official_absmax": 0.609375 }, "model.language_model.layers.47.post_attention_layernorm.weight": { "numel": 5120, "differing": 1580, "frac": 0.30859, "max_abs": 0.00048828125, "rel_l1": 0.0008238453883677721, "official_absmax": 0.66796875 }, "model.language_model.layers.47.self_attn.k_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00018310546875, "rel_l1": 1.0835915418283548e-06, "official_absmax": 2.21875 }, "model.language_model.layers.47.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.0001220703125, "rel_l1": 7.300916990971018e-07, "official_absmax": 1.3984375 }, "model.language_model.layers.48.input_layernorm.weight": { "numel": 5120, "differing": 2098, "frac": 0.40977, "max_abs": 0.00048828125, "rel_l1": 0.001363830640912056, "official_absmax": 0.6015625 }, "model.language_model.layers.48.linear_attn.conv1d.weight": { "numel": 40960, "differing": 44, "frac": 0.00107, "max_abs": 2.9802322387695312e-08, "rel_l1": 9.026039360371385e-10, "official_absmax": 0.875 }, "model.language_model.layers.48.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.4375 }, "model.language_model.layers.48.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.21875 }, "model.language_model.layers.48.post_attention_layernorm.weight": { "numel": 5120, "differing": 1794, "frac": 0.35039, "max_abs": 0.00048828125, "rel_l1": 0.0010481268400326371, "official_absmax": 0.66796875 }, "model.language_model.layers.49.input_layernorm.weight": { "numel": 5120, "differing": 2022, "frac": 0.39492, "max_abs": 0.00048828125, "rel_l1": 0.0012655318714678288, "official_absmax": 0.59375 }, "model.language_model.layers.49.linear_attn.conv1d.weight": { "numel": 40960, "differing": 50, "frac": 0.00122, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1352786488671995e-09, "official_absmax": 0.765625 }, "model.language_model.layers.49.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.46875 }, "model.language_model.layers.49.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.359375 }, "model.language_model.layers.49.post_attention_layernorm.weight": { "numel": 5120, "differing": 1949, "frac": 0.38066, "max_abs": 0.00048828125, "rel_l1": 0.0013363775797188282, "official_absmax": 0.59375 }, "model.language_model.layers.5.input_layernorm.weight": { "numel": 5120, "differing": 2647, "frac": 0.51699, "max_abs": 0.00048828125, "rel_l1": 0.0023074268829077482, "official_absmax": 0.61328125 }, "model.language_model.layers.5.linear_attn.conv1d.weight": { "numel": 40960, "differing": 39, "frac": 0.00095, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.3613616900443049e-09, "official_absmax": 0.474609375 }, "model.language_model.layers.5.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.28125 }, "model.language_model.layers.5.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0078125 }, "model.language_model.layers.5.post_attention_layernorm.weight": { "numel": 5120, "differing": 1526, "frac": 0.29805, "max_abs": 0.000244140625, "rel_l1": 0.0008383318781852722, "official_absmax": 0.99609375 }, "model.language_model.layers.50.input_layernorm.weight": { "numel": 5120, "differing": 2529, "frac": 0.49395, "max_abs": 0.00048828125, "rel_l1": 0.001983830239623785, "official_absmax": 0.5078125 }, "model.language_model.layers.50.linear_attn.conv1d.weight": { "numel": 40960, "differing": 36, "frac": 0.00088, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.276433344749478e-10, "official_absmax": 0.67578125 }, "model.language_model.layers.50.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.71875 }, "model.language_model.layers.50.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.1328125 }, "model.language_model.layers.50.post_attention_layernorm.weight": { "numel": 5120, "differing": 2425, "frac": 0.47363, "max_abs": 0.00048828125, "rel_l1": 0.0019366717897355556, "official_absmax": 0.455078125 }, "model.language_model.layers.51.input_layernorm.weight": { "numel": 5120, "differing": 3032, "frac": 0.59219, "max_abs": 0.00048828125, "rel_l1": 0.003025392070412636, "official_absmax": 0.51953125 }, "model.language_model.layers.51.post_attention_layernorm.weight": { "numel": 5120, "differing": 1722, "frac": 0.33633, "max_abs": 0.00048828125, "rel_l1": 0.0011294144205749035, "official_absmax": 0.59375 }, "model.language_model.layers.51.self_attn.k_norm.weight": { "numel": 256, "differing": 4, "frac": 0.01562, "max_abs": 0.00048828125, "rel_l1": 8.056208571360912e-06, "official_absmax": 1.703125 }, "model.language_model.layers.51.self_attn.q_norm.weight": { "numel": 256, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.359375 }, "model.language_model.layers.52.input_layernorm.weight": { "numel": 5120, "differing": 860, "frac": 0.16797, "max_abs": 0.00048828125, "rel_l1": 0.0004561996611300856, "official_absmax": 0.5078125 }, "model.language_model.layers.52.linear_attn.conv1d.weight": { "numel": 40960, "differing": 41, "frac": 0.001, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.582982409713111e-10, "official_absmax": 0.87109375 }, "model.language_model.layers.52.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.9375 }, "model.language_model.layers.52.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.015625 }, "model.language_model.layers.52.post_attention_layernorm.weight": { "numel": 5120, "differing": 1199, "frac": 0.23418, "max_abs": 0.00048828125, "rel_l1": 0.0005771254654973745, "official_absmax": 0.71484375 }, "model.language_model.layers.53.input_layernorm.weight": { "numel": 5120, "differing": 971, "frac": 0.18965, "max_abs": 0.00048828125, "rel_l1": 0.0004906042013317347, "official_absmax": 0.498046875 }, "model.language_model.layers.53.linear_attn.conv1d.weight": { "numel": 40960, "differing": 29, "frac": 0.00071, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.09289130365309e-10, "official_absmax": 0.6328125 }, "model.language_model.layers.53.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.6875 }, "model.language_model.layers.53.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0078125 }, "model.language_model.layers.53.post_attention_layernorm.weight": { "numel": 5120, "differing": 1076, "frac": 0.21016, "max_abs": 0.00048828125, "rel_l1": 0.0004697846597991884, "official_absmax": 0.63671875 }, "model.language_model.layers.54.input_layernorm.weight": { "numel": 5120, "differing": 1087, "frac": 0.2123, "max_abs": 0.00048828125, "rel_l1": 0.0006181440548971295, "official_absmax": 0.640625 }, "model.language_model.layers.54.linear_attn.conv1d.weight": { "numel": 40960, "differing": 22, "frac": 0.00054, "max_abs": 2.9802322387695312e-08, "rel_l1": 3.6539254955059164e-10, "official_absmax": 0.54296875 }, "model.language_model.layers.54.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.40625 }, "model.language_model.layers.54.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0625 }, "model.language_model.layers.54.post_attention_layernorm.weight": { "numel": 5120, "differing": 741, "frac": 0.14473, "max_abs": 0.00048828125, "rel_l1": 0.00026078199152834713, "official_absmax": 0.54296875 }, "model.language_model.layers.55.input_layernorm.weight": { "numel": 5120, "differing": 2693, "frac": 0.52598, "max_abs": 0.00048828125, "rel_l1": 0.0022114599123597145, "official_absmax": 0.8046875 }, "model.language_model.layers.55.post_attention_layernorm.weight": { "numel": 5120, "differing": 594, "frac": 0.11602, "max_abs": 0.00048828125, "rel_l1": 0.00023812630388420075, "official_absmax": 0.63671875 }, "model.language_model.layers.55.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.000244140625, "rel_l1": 3.444283947828808e-06, "official_absmax": 1.734375 }, "model.language_model.layers.55.self_attn.q_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 4.271292254998116e-06, "official_absmax": 1.2890625 }, "model.language_model.layers.56.input_layernorm.weight": { "numel": 5120, "differing": 624, "frac": 0.12187, "max_abs": 0.00048828125, "rel_l1": 0.0003515674325171858, "official_absmax": 0.765625 }, "model.language_model.layers.56.linear_attn.conv1d.weight": { "numel": 40960, "differing": 20, "frac": 0.00049, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.6951463283353405e-10, "official_absmax": 0.7421875 }, "model.language_model.layers.56.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.90625 }, "model.language_model.layers.56.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0234375 }, "model.language_model.layers.56.post_attention_layernorm.weight": { "numel": 5120, "differing": 406, "frac": 0.0793, "max_abs": 0.00048828125, "rel_l1": 0.00012710748706012964, "official_absmax": 0.65234375 }, "model.language_model.layers.57.input_layernorm.weight": { "numel": 5120, "differing": 535, "frac": 0.10449, "max_abs": 0.00048828125, "rel_l1": 0.0002792677260003984, "official_absmax": 0.9609375 }, "model.language_model.layers.57.linear_attn.conv1d.weight": { "numel": 40960, "differing": 20, "frac": 0.00049, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.7749555431277884e-10, "official_absmax": 0.65234375 }, "model.language_model.layers.57.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.15625 }, "model.language_model.layers.57.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.4921875 }, "model.language_model.layers.57.post_attention_layernorm.weight": { "numel": 5120, "differing": 261, "frac": 0.05098, "max_abs": 0.00048828125, "rel_l1": 6.271326128626242e-05, "official_absmax": 0.7890625 }, "model.language_model.layers.58.input_layernorm.weight": { "numel": 5120, "differing": 515, "frac": 0.10059, "max_abs": 0.00048828125, "rel_l1": 0.00026316760340705514, "official_absmax": 1.03125 }, "model.language_model.layers.58.linear_attn.conv1d.weight": { "numel": 40960, "differing": 22, "frac": 0.00054, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.9735450213230763e-10, "official_absmax": 0.62109375 }, "model.language_model.layers.58.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.125 }, "model.language_model.layers.58.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.25 }, "model.language_model.layers.58.post_attention_layernorm.weight": { "numel": 5120, "differing": 181, "frac": 0.03535, "max_abs": 0.00048828125, "rel_l1": 3.772572017624043e-05, "official_absmax": 0.83984375 }, "model.language_model.layers.59.input_layernorm.weight": { "numel": 5120, "differing": 2420, "frac": 0.47266, "max_abs": 0.00048828125, "rel_l1": 0.0013005259679630399, "official_absmax": 1.0078125 }, "model.language_model.layers.59.post_attention_layernorm.weight": { "numel": 5120, "differing": 102, "frac": 0.01992, "max_abs": 0.00048828125, "rel_l1": 2.4727003619773313e-05, "official_absmax": 0.77734375 }, "model.language_model.layers.59.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 6.076805675547803e-06, "official_absmax": 1.609375 }, "model.language_model.layers.59.self_attn.q_norm.weight": { "numel": 256, "differing": 1, "frac": 0.00391, "max_abs": 0.00048828125, "rel_l1": 3.5188593301427318e-06, "official_absmax": 1.328125 }, "model.language_model.layers.6.input_layernorm.weight": { "numel": 5120, "differing": 2925, "frac": 0.57129, "max_abs": 0.00048828125, "rel_l1": 0.0027903385926038027, "official_absmax": 0.4296875 }, "model.language_model.layers.6.linear_attn.conv1d.weight": { "numel": 40960, "differing": 15, "frac": 0.00037, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.7123350494237e-10, "official_absmax": 0.5078125 }, "model.language_model.layers.6.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.6875 }, "model.language_model.layers.6.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.0 }, "model.language_model.layers.6.post_attention_layernorm.weight": { "numel": 5120, "differing": 970, "frac": 0.18945, "max_abs": 0.00048828125, "rel_l1": 0.0004960015066899359, "official_absmax": 0.87890625 }, "model.language_model.layers.60.input_layernorm.weight": { "numel": 5120, "differing": 26, "frac": 0.00508, "max_abs": 0.00048828125, "rel_l1": 1.0377670150774065e-05, "official_absmax": 1.328125 }, "model.language_model.layers.60.linear_attn.conv1d.weight": { "numel": 40960, "differing": 21, "frac": 0.00051, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.361794648741977e-10, "official_absmax": 0.74609375 }, "model.language_model.layers.60.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.9375 }, "model.language_model.layers.60.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.265625 }, "model.language_model.layers.60.post_attention_layernorm.weight": { "numel": 5120, "differing": 18, "frac": 0.00352, "max_abs": 0.00048828125, "rel_l1": 4.494042059377534e-06, "official_absmax": 0.67578125 }, "model.language_model.layers.61.input_layernorm.weight": { "numel": 5120, "differing": 549, "frac": 0.10723, "max_abs": 0.00048828125, "rel_l1": 0.00037051463732495904, "official_absmax": 1.1796875 }, "model.language_model.layers.61.linear_attn.conv1d.weight": { "numel": 40960, "differing": 23, "frac": 0.00056, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.736637860767388e-10, "official_absmax": 0.62890625 }, "model.language_model.layers.61.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.875 }, "model.language_model.layers.61.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.390625 }, "model.language_model.layers.61.post_attention_layernorm.weight": { "numel": 5120, "differing": 17, "frac": 0.00332, "max_abs": 0.00048828125, "rel_l1": 3.140223498121486e-06, "official_absmax": 0.64453125 }, "model.language_model.layers.62.input_layernorm.weight": { "numel": 5120, "differing": 986, "frac": 0.19258, "max_abs": 0.00048828125, "rel_l1": 0.0006647807313129306, "official_absmax": 1.1328125 }, "model.language_model.layers.62.linear_attn.conv1d.weight": { "numel": 40960, "differing": 32, "frac": 0.00078, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.3022632678635375e-10, "official_absmax": 0.63671875 }, "model.language_model.layers.62.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.0625 }, "model.language_model.layers.62.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.390625 }, "model.language_model.layers.62.post_attention_layernorm.weight": { "numel": 5120, "differing": 19, "frac": 0.00371, "max_abs": 0.00048828125, "rel_l1": 3.7307975162548246e-06, "official_absmax": 0.61328125 }, "model.language_model.layers.63.input_layernorm.weight": { "numel": 5120, "differing": 2453, "frac": 0.4791, "max_abs": 0.00048828125, "rel_l1": 0.0014652158133685589, "official_absmax": 1.015625 }, "model.language_model.layers.63.post_attention_layernorm.weight": { "numel": 5120, "differing": 131, "frac": 0.02559, "max_abs": 0.00048828125, "rel_l1": 4.868942414759658e-05, "official_absmax": 0.515625 }, "model.language_model.layers.63.self_attn.k_norm.weight": { "numel": 256, "differing": 5, "frac": 0.01953, "max_abs": 0.00048828125, "rel_l1": 1.4954332982597407e-05, "official_absmax": 1.609375 }, "model.language_model.layers.63.self_attn.k_proj.weight": { "numel": 5242880, "differing": 2001, "frac": 0.00038, "max_abs": 2.9802322387695312e-08, "rel_l1": 8.458854172666008e-10, "official_absmax": 0.271484375 }, "model.language_model.layers.63.self_attn.q_norm.weight": { "numel": 256, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.2109375 }, "model.language_model.layers.63.self_attn.q_proj.weight": { "numel": 62914560, "differing": 18994, "frac": 0.0003, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.079045045126463e-10, "official_absmax": 0.5390625 }, "model.language_model.layers.63.self_attn.v_proj.weight": { "numel": 5242880, "differing": 943, "frac": 0.00018, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.0475943163233978e-10, "official_absmax": 1.015625 }, "model.language_model.layers.7.input_layernorm.weight": { "numel": 5120, "differing": 2863, "frac": 0.55918, "max_abs": 0.00048828125, "rel_l1": 0.0017932679038494825, "official_absmax": 1.0546875 }, "model.language_model.layers.7.post_attention_layernorm.weight": { "numel": 5120, "differing": 3251, "frac": 0.63496, "max_abs": 0.00048828125, "rel_l1": 0.0019841245375573635, "official_absmax": 1.0 }, "model.language_model.layers.7.self_attn.k_norm.weight": { "numel": 256, "differing": 5, "frac": 0.01953, "max_abs": 0.00048828125, "rel_l1": 1.753490141709335e-05, "official_absmax": 0.84375 }, "model.language_model.layers.7.self_attn.q_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 1.1333242582622916e-05, "official_absmax": 0.515625 }, "model.language_model.layers.8.input_layernorm.weight": { "numel": 5120, "differing": 2468, "frac": 0.48203, "max_abs": 0.00048828125, "rel_l1": 0.0022696792148053646, "official_absmax": 0.578125 }, "model.language_model.layers.8.linear_attn.conv1d.weight": { "numel": 40960, "differing": 18, "frac": 0.00044, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.849773443422635e-10, "official_absmax": 0.40234375 }, "model.language_model.layers.8.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.59375 }, "model.language_model.layers.8.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.99609375 }, "model.language_model.layers.8.post_attention_layernorm.weight": { "numel": 5120, "differing": 3285, "frac": 0.6416, "max_abs": 0.00048828125, "rel_l1": 0.0021054481621831656, "official_absmax": 0.99609375 }, "model.language_model.layers.9.input_layernorm.weight": { "numel": 5120, "differing": 2806, "frac": 0.54805, "max_abs": 0.00048828125, "rel_l1": 0.0026783186476677656, "official_absmax": 0.55859375 }, "model.language_model.layers.9.linear_attn.conv1d.weight": { "numel": 40960, "differing": 21, "frac": 0.00051, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.287357939311789e-10, "official_absmax": 0.490234375 }, "model.language_model.layers.9.linear_attn.dt_bias": { "numel": 48, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.71875 }, "model.language_model.layers.9.linear_attn.norm.weight": { "numel": 128, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.9765625 }, "model.language_model.layers.9.post_attention_layernorm.weight": { "numel": 5120, "differing": 2955, "frac": 0.57715, "max_abs": 0.00048828125, "rel_l1": 0.001750895637087524, "official_absmax": 0.99609375 }, "model.language_model.norm.weight": { "numel": 5120, "differing": 5, "frac": 0.00098, "max_abs": 0.00048828125, "rel_l1": 3.345437278312602e-07, "official_absmax": 1.7109375 }, "model.visual.blocks.0.attn.proj.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.578125 }, "model.visual.blocks.0.attn.proj.weight": { "numel": 1327104, "differing": 651, "frac": 0.00049, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.2522900494360556e-09, "official_absmax": 0.302734375 }, "model.visual.blocks.0.attn.qkv.bias": { "numel": 3456, "differing": 1, "frac": 0.00029, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.0570758638149833e-11, "official_absmax": 6.1875 }, "model.visual.blocks.0.attn.qkv.weight": { "numel": 3981312, "differing": 2414, "frac": 0.00061, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1725603821233221e-09, "official_absmax": 0.58203125 }, "model.visual.blocks.0.mlp.linear_fc1.bias": { "numel": 4304, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 9.4375 }, "model.visual.blocks.0.mlp.linear_fc1.weight": { "numel": 4958208, "differing": 15543, "frac": 0.00313, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.434245833209502e-09, "official_absmax": 0.388671875 }, "model.visual.blocks.0.mlp.linear_fc2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 11.25 }, "model.visual.blocks.0.mlp.linear_fc2.weight": { "numel": 4958208, "differing": 14935, "frac": 0.00301, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.783399761201565e-09, "official_absmax": 0.224609375 }, "model.visual.blocks.0.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.921875 }, "model.visual.blocks.0.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.140625 }, "model.visual.blocks.0.norm2.bias": { "numel": 1152, "differing": 3, "frac": 0.0026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.224189609214335e-10, "official_absmax": 3.84375 }, "model.visual.blocks.0.norm2.weight": { "numel": 1152, "differing": 1, "frac": 0.00087, "max_abs": 2.0489096641540527e-08, "rel_l1": 9.25992182237989e-12, "official_absmax": 14.5625 }, "model.visual.blocks.1.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.78125 }, "model.visual.blocks.1.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.0625 }, "model.visual.blocks.1.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.71875 }, "model.visual.blocks.1.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.703125 }, "model.visual.blocks.10.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.34375 }, "model.visual.blocks.10.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.71875 }, "model.visual.blocks.10.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.609375 }, "model.visual.blocks.10.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.46875 }, "model.visual.blocks.11.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.28125 }, "model.visual.blocks.11.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.828125 }, "model.visual.blocks.11.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.203125 }, "model.visual.blocks.11.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.40625 }, "model.visual.blocks.12.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.21875 }, "model.visual.blocks.12.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.6875 }, "model.visual.blocks.12.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.21875 }, "model.visual.blocks.12.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.6875 }, "model.visual.blocks.13.attn.proj.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.453125 }, "model.visual.blocks.13.attn.proj.weight": { "numel": 1327104, "differing": 402, "frac": 0.0003, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.158695775582146e-10, "official_absmax": 0.13671875 }, "model.visual.blocks.13.attn.qkv.bias": { "numel": 3456, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.0625 }, "model.visual.blocks.13.attn.qkv.weight": { "numel": 3981312, "differing": 1073, "frac": 0.00027, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.735233316566223e-10, "official_absmax": 0.2080078125 }, "model.visual.blocks.13.mlp.linear_fc1.bias": { "numel": 4304, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.515625 }, "model.visual.blocks.13.mlp.linear_fc1.weight": { "numel": 4958208, "differing": 1274, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.612332737963243e-10, "official_absmax": 0.1904296875 }, "model.visual.blocks.13.mlp.linear_fc2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.390625 }, "model.visual.blocks.13.mlp.linear_fc2.weight": { "numel": 4958208, "differing": 1524, "frac": 0.00031, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.221754222934806e-10, "official_absmax": 0.375 }, "model.visual.blocks.13.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.71875 }, "model.visual.blocks.13.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.6875 }, "model.visual.blocks.13.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.890625 }, "model.visual.blocks.13.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.421875 }, "model.visual.blocks.14.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.1875 }, "model.visual.blocks.14.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.0625 }, "model.visual.blocks.14.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.9375 }, "model.visual.blocks.14.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.4375 }, "model.visual.blocks.15.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.625 }, "model.visual.blocks.15.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.125 }, "model.visual.blocks.15.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.28125 }, "model.visual.blocks.15.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.3125 }, "model.visual.blocks.16.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 11.375 }, "model.visual.blocks.16.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.9921875 }, "model.visual.blocks.16.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.40625 }, "model.visual.blocks.16.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.34375 }, "model.visual.blocks.17.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 14.0625 }, "model.visual.blocks.17.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.8046875 }, "model.visual.blocks.17.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 11.375 }, "model.visual.blocks.17.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.59375 }, "model.visual.blocks.18.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 12.375 }, "model.visual.blocks.18.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.875 }, "model.visual.blocks.18.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 5.5625 }, "model.visual.blocks.18.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.0 }, "model.visual.blocks.19.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 15.25 }, "model.visual.blocks.19.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.8125 }, "model.visual.blocks.19.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 8.625 }, "model.visual.blocks.19.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 9.25 }, "model.visual.blocks.2.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.65625 }, "model.visual.blocks.2.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.15625 }, "model.visual.blocks.2.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.984375 }, "model.visual.blocks.2.norm2.weight": { "numel": 1152, "differing": 1, "frac": 0.00087, "max_abs": 1.4901161193847656e-08, "rel_l1": 1.2736842830429751e-11, "official_absmax": 2.5 }, "model.visual.blocks.20.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 14.25 }, "model.visual.blocks.20.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.8359375 }, "model.visual.blocks.20.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 25.5 }, "model.visual.blocks.20.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 10.125 }, "model.visual.blocks.21.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 12.25 }, "model.visual.blocks.21.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.8828125 }, "model.visual.blocks.21.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 9.1875 }, "model.visual.blocks.21.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 11.75 }, "model.visual.blocks.22.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 9.1875 }, "model.visual.blocks.22.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.09375 }, "model.visual.blocks.22.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 9.1875 }, "model.visual.blocks.22.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 14.125 }, "model.visual.blocks.23.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 10.0 }, "model.visual.blocks.23.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.140625 }, "model.visual.blocks.23.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 10.1875 }, "model.visual.blocks.23.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 15.6875 }, "model.visual.blocks.24.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.90625 }, "model.visual.blocks.24.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.28125 }, "model.visual.blocks.24.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 14.0 }, "model.visual.blocks.24.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 17.5 }, "model.visual.blocks.25.norm1.bias": { "numel": 1152, "differing": 1, "frac": 0.00087, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.1485704748404046e-10, "official_absmax": 6.96875 }, "model.visual.blocks.25.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.5625 }, "model.visual.blocks.25.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 20.125 }, "model.visual.blocks.25.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 12.5 }, "model.visual.blocks.26.attn.proj.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 7.09375 }, "model.visual.blocks.26.attn.proj.weight": { "numel": 1327104, "differing": 344, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 3.986091734464736e-10, "official_absmax": 0.369140625 }, "model.visual.blocks.26.attn.qkv.bias": { "numel": 3456, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.96875 }, "model.visual.blocks.26.attn.qkv.weight": { "numel": 3981312, "differing": 1061, "frac": 0.00027, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.562616395809016e-10, "official_absmax": 0.1572265625 }, "model.visual.blocks.26.mlp.linear_fc1.bias": { "numel": 4304, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.15625 }, "model.visual.blocks.26.mlp.linear_fc1.weight": { "numel": 4958208, "differing": 1280, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.2418346613004587e-10, "official_absmax": 0.248046875 }, "model.visual.blocks.26.mlp.linear_fc2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 23.375 }, "model.visual.blocks.26.mlp.linear_fc2.weight": { "numel": 4958208, "differing": 1282, "frac": 0.00026, "max_abs": 2.9802322387695312e-08, "rel_l1": 4.2014999812600706e-10, "official_absmax": 1.109375 }, "model.visual.blocks.26.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 6.96875 }, "model.visual.blocks.26.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.546875 }, "model.visual.blocks.26.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 20.125 }, "model.visual.blocks.26.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 15.75 }, "model.visual.blocks.3.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.65625 }, "model.visual.blocks.3.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.875 }, "model.visual.blocks.3.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.59375 }, "model.visual.blocks.3.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.046875 }, "model.visual.blocks.4.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.4375 }, "model.visual.blocks.4.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.984375 }, "model.visual.blocks.4.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.84375 }, "model.visual.blocks.4.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.9296875 }, "model.visual.blocks.5.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.09375 }, "model.visual.blocks.5.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.5703125 }, "model.visual.blocks.5.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.515625 }, "model.visual.blocks.5.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.140625 }, "model.visual.blocks.6.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.515625 }, "model.visual.blocks.6.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.484375 }, "model.visual.blocks.6.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 4.3125 }, "model.visual.blocks.6.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.234375 }, "model.visual.blocks.7.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.3125 }, "model.visual.blocks.7.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.40625 }, "model.visual.blocks.7.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.84375 }, "model.visual.blocks.7.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.09375 }, "model.visual.blocks.8.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.3125 }, "model.visual.blocks.8.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.5 }, "model.visual.blocks.8.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.9375 }, "model.visual.blocks.8.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.3125 }, "model.visual.blocks.9.norm1.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.15625 }, "model.visual.blocks.9.norm1.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.375 }, "model.visual.blocks.9.norm2.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.34375 }, "model.visual.blocks.9.norm2.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.640625 }, "model.visual.merger.linear_fc1.bias": { "numel": 4608, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.3515625 }, "model.visual.merger.linear_fc1.weight": { "numel": 21233664, "differing": 6468, "frac": 0.0003, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.447629097294794e-10, "official_absmax": 0.181640625 }, "model.visual.merger.linear_fc2.bias": { "numel": 5120, "differing": 3, "frac": 0.00059, "max_abs": 2.9802322387695312e-08, "rel_l1": 2.171700153041911e-09, "official_absmax": 0.75390625 }, "model.visual.merger.linear_fc2.weight": { "numel": 23592960, "differing": 7713, "frac": 0.00033, "max_abs": 2.9802322387695312e-08, "rel_l1": 7.399645340910865e-10, "official_absmax": 0.47265625 }, "model.visual.merger.norm.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 11.9375 }, "model.visual.merger.norm.weight": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 2.421875 }, "model.visual.patch_embed.proj.bias": { "numel": 1152, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 3.046875 }, "model.visual.patch_embed.proj.weight": { "numel": 1769472, "differing": 2042, "frac": 0.00115, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.588306262183096e-09, "official_absmax": 0.0751953125 }, "model.visual.pos_embed.weight": { "numel": 2654208, "differing": 376, "frac": 0.00014, "max_abs": 2.9802322387695312e-08, "rel_l1": 8.19744966351621e-11, "official_absmax": 7.03125 }, "mtp.fc.weight": { "numel": 52428800, "differing": 21383, "frac": 0.00041, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.0922971416249538e-09, "official_absmax": 0.9140625 }, "mtp.layers.0.input_layernorm.weight": { "numel": 5120, "differing": 2631, "frac": 0.51387, "max_abs": 0.00048828125, "rel_l1": 0.001564409933052957, "official_absmax": 0.953125 }, "mtp.layers.0.mlp.down_proj.weight": { "numel": 89128960, "differing": 67813850, "frac": 0.76085, "max_abs": 0.00341796875, "rel_l1": 0.00604272773489356, "official_absmax": 0.8828125 }, "mtp.layers.0.mlp.gate_proj.weight": { "numel": 89128960, "differing": 66326033, "frac": 0.74416, "max_abs": 0.00146484375, "rel_l1": 0.005541860591620207, "official_absmax": 0.34765625 }, "mtp.layers.0.mlp.up_proj.weight": { "numel": 89128960, "differing": 66372734, "frac": 0.74468, "max_abs": 0.00146484375, "rel_l1": 0.005552587099373341, "official_absmax": 0.36328125 }, "mtp.layers.0.post_attention_layernorm.weight": { "numel": 5120, "differing": 196, "frac": 0.03828, "max_abs": 0.00048828125, "rel_l1": 8.555685053579509e-05, "official_absmax": 0.515625 }, "mtp.layers.0.self_attn.k_norm.weight": { "numel": 256, "differing": 3, "frac": 0.01172, "max_abs": 0.00048828125, "rel_l1": 5.499094640981639e-06, "official_absmax": 1.7890625 }, "mtp.layers.0.self_attn.k_proj.weight": { "numel": 5242880, "differing": 1564, "frac": 0.0003, "max_abs": 2.9802322387695312e-08, "rel_l1": 5.581485273964404e-10, "official_absmax": 0.2734375 }, "mtp.layers.0.self_attn.o_proj.weight": { "numel": 31457280, "differing": 24006059, "frac": 0.76313, "max_abs": 0.0048828125, "rel_l1": 0.006134907249361277, "official_absmax": 1.2109375 }, "mtp.layers.0.self_attn.q_norm.weight": { "numel": 256, "differing": 2, "frac": 0.00781, "max_abs": 0.00048828125, "rel_l1": 4.784231350640766e-06, "official_absmax": 1.96875 }, "mtp.layers.0.self_attn.q_proj.weight": { "numel": 62914560, "differing": 21058, "frac": 0.00033, "max_abs": 2.9802322387695312e-08, "rel_l1": 6.494492166275734e-10, "official_absmax": 0.474609375 }, "mtp.layers.0.self_attn.v_proj.weight": { "numel": 5242880, "differing": 782, "frac": 0.00015, "max_abs": 2.9802322387695312e-08, "rel_l1": 1.454596304517608e-10, "official_absmax": 0.53125 }, "mtp.norm.weight": { "numel": 5120, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 1.9296875 }, "mtp.pre_fc_norm_embedding.weight": { "numel": 5120, "differing": 0, "frac": 0.0, "max_abs": 0.0, "rel_l1": 0.0, "official_absmax": 0.75 }, "mtp.pre_fc_norm_hidden.weight": { "numel": 5120, "differing": 180, "frac": 0.03516, "max_abs": 0.00048828125, "rel_l1": 5.457092265714891e-05, "official_absmax": 0.455078125 } }