Files
WarpQuant-Llama-3-8B-R16E4H4/tqi3_fake_quant.json
ModelHub XC 3bd3e1f70b 初始化项目,由ModelHub XC社区提供模型
Model: HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4
Source: Original Platform
2026-08-31 09:33:18 +08:00

3152 lines
86 KiB
JSON

{
"activations": "/workspace/llama3-warpquant/activations",
"created_at": "2026-08-14T08:25:26.827653+00:00",
"damping": 0.5,
"fake_quantized": true,
"gptq_error_feedback": true,
"group_size": 128,
"promotion_fraction": 0.3671875,
"quantized_tensors": 224,
"seed": 1414613299,
"selection_manifest": "/workspace/llama3-warpquant/curvature/manifest.json",
"source": "/workspace/models/Meta-Llama-3-8B",
"tensors": {
"model.layers.0.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.015937710513013792,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.3308872070047073,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.7031736380013172,
"shape": [
4096,
14336
]
},
"model.layers.0.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.03559036074620749,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014705806010169908,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1303548809955828,
"shape": [
14336,
4096
]
},
"model.layers.0.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04200425613532713,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014716545003466308,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1267571240023244,
"shape": [
14336,
4096
]
},
"model.layers.0.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.009958144834181238,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014551027998095378,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13680504000512883,
"shape": [
1024,
4096
]
},
"model.layers.0.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.03681738242977615,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014612491009756923,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3196648209996056,
"shape": [
4096,
4096
]
},
"model.layers.0.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.00835441430441703,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014593176980270073,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3183482550084591,
"shape": [
4096,
4096
]
},
"model.layers.0.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.10468198465667863,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014669297990621999,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13437573402188718,
"shape": [
1024,
4096
]
},
"model.layers.1.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.014680927698630778,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26538383300066926,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4253008590021636,
"shape": [
4096,
14336
]
},
"model.layers.1.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.0411034789351933,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014604886993765831,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.132027621002635,
"shape": [
14336,
4096
]
},
"model.layers.1.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04570369387677158,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014604066993342713,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1195683300029486,
"shape": [
14336,
4096
]
},
"model.layers.1.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.009381644225409788,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01461602799827233,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12916735099861398,
"shape": [
1024,
4096
]
},
"model.layers.1.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.03716823132461608,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014614205021644011,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31968017702456564,
"shape": [
4096,
4096
]
},
"model.layers.1.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.009616522143756379,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014600380003685132,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31648389098700136,
"shape": [
4096,
4096
]
},
"model.layers.1.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.08303001983349607,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014615928987041116,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13493670400930569,
"shape": [
1024,
4096
]
},
"model.layers.10.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.0320130218049811,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.27319980098400265,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4307553590042517,
"shape": [
4096,
14336
]
},
"model.layers.10.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.025833007259724018,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01465235001523979,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1304956820094958,
"shape": [
14336,
4096
]
},
"model.layers.10.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04285121192110673,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014648793003289029,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1322728910017759,
"shape": [
14336,
4096
]
},
"model.layers.10.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.019820162420173325,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014657179010100663,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12939836099394597,
"shape": [
1024,
4096
]
},
"model.layers.10.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.0633724476158651,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014655946986749768,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.322024719993351,
"shape": [
4096,
4096
]
},
"model.layers.10.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.022117404125098574,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014664793008705601,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3190295660169795,
"shape": [
4096,
4096
]
},
"model.layers.10.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04636030251080129,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01462997798807919,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1299592360155657,
"shape": [
1024,
4096
]
},
"model.layers.11.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028322392778659104,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2677451900090091,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4209276600158773,
"shape": [
4096,
14336
]
},
"model.layers.11.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02600439966445865,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014636910986155272,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1346062239899766,
"shape": [
14336,
4096
]
},
"model.layers.11.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04307867317179749,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014638104010373354,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1355084509996232,
"shape": [
14336,
4096
]
},
"model.layers.11.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020267691876366536,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014647180010797456,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12835646199528128,
"shape": [
1024,
4096
]
},
"model.layers.11.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.06668723877911359,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014595464017475024,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3210202640038915,
"shape": [
4096,
4096
]
},
"model.layers.11.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02161551240085763,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014673079014755785,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3156259010138456,
"shape": [
4096,
4096
]
},
"model.layers.11.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04424530932212295,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.0146348679845687,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1285538400115911,
"shape": [
1024,
4096
]
},
"model.layers.12.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.027404714527101273,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26968229899648577,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4221025859878864,
"shape": [
4096,
14336
]
},
"model.layers.12.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02546071924490612,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014617475011618808,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1277417480014265,
"shape": [
14336,
4096
]
},
"model.layers.12.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04186881223061151,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014590003993362188,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1228092389937956,
"shape": [
14336,
4096
]
},
"model.layers.12.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.0174058534043335,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014683308021631092,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1293022689933423,
"shape": [
1024,
4096
]
},
"model.layers.12.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.057922612287355325,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014649955002823845,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3204046339960769,
"shape": [
4096,
4096
]
},
"model.layers.12.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.01890739005824379,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014641621004557237,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31539562001125887,
"shape": [
4096,
4096
]
},
"model.layers.12.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04813920931457681,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014597917994251475,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1282280260056723,
"shape": [
1024,
4096
]
},
"model.layers.13.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028843241516937977,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26553816001978703,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4119335069844965,
"shape": [
4096,
14336
]
},
"model.layers.13.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.024217616205661045,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01464169000973925,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1346196130034514,
"shape": [
14336,
4096
]
},
"model.layers.13.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.041912997013580124,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01465222000842914,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1185657279856969,
"shape": [
14336,
4096
]
},
"model.layers.13.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020338448734831125,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014605211996240541,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12810219399398193,
"shape": [
1024,
4096
]
},
"model.layers.13.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.055430245282634245,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014632764999987558,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3209068509750068,
"shape": [
4096,
4096
]
},
"model.layers.13.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.023807349351640274,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014639526983955875,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3156330190249719,
"shape": [
4096,
4096
]
},
"model.layers.13.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04980014744585993,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014597510016756132,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13038333697477356,
"shape": [
1024,
4096
]
},
"model.layers.14.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.03080083790807457,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2662895679823123,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4153816390025895,
"shape": [
4096,
14336
]
},
"model.layers.14.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.023375603248083552,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014647403004346415,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1280329760047607,
"shape": [
14336,
4096
]
},
"model.layers.14.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04394749608005593,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014612016006140038,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1233870539872441,
"shape": [
14336,
4096
]
},
"model.layers.14.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.019331503210441865,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014637804997619241,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12827536000986584,
"shape": [
1024,
4096
]
},
"model.layers.14.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.059439730925794625,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014615463005611673,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3203331279801205,
"shape": [
4096,
4096
]
},
"model.layers.14.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.023223435504152956,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014675274025648832,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3166496330231894,
"shape": [
4096,
4096
]
},
"model.layers.14.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04792568028217769,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014631781989010051,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12871611199807376,
"shape": [
1024,
4096
]
},
"model.layers.15.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.030107530996912876,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2736986359814182,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.420295124000404,
"shape": [
4096,
14336
]
},
"model.layers.15.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.022319837837168895,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014664974994957447,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1309605209971778,
"shape": [
14336,
4096
]
},
"model.layers.15.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04378244598506008,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014661769004305825,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1251995689817704,
"shape": [
14336,
4096
]
},
"model.layers.15.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.0196738845693603,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014637003012467176,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12889245300902985,
"shape": [
1024,
4096
]
},
"model.layers.15.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.061176718576124285,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014527989987982437,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3198720369837247,
"shape": [
4096,
4096
]
},
"model.layers.15.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.021072374811165123,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014641892979852855,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3156102130014915,
"shape": [
4096,
4096
]
},
"model.layers.15.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.051605684480451726,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014644908020272851,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12818663800135255,
"shape": [
1024,
4096
]
},
"model.layers.16.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.03060384874677205,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.25921246100915596,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4054513289884198,
"shape": [
4096,
14336
]
},
"model.layers.16.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.023675509860478998,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014648915006546304,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1260714259988163,
"shape": [
14336,
4096
]
},
"model.layers.16.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.044021549626707256,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01464243300142698,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1251569179876242,
"shape": [
14336,
4096
]
},
"model.layers.16.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.019344081160406094,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014707455004099756,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1297916829935275,
"shape": [
1024,
4096
]
},
"model.layers.16.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.051095062660552346,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014520986995194107,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3210381690005306,
"shape": [
4096,
4096
]
},
"model.layers.16.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.022836303231756807,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014655920007498935,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31609576698974706,
"shape": [
4096,
4096
]
},
"model.layers.16.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.05076256430723606,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014701703999890015,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13260309299221262,
"shape": [
1024,
4096
]
},
"model.layers.17.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028658251715113433,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2661531350167934,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4179458139988128,
"shape": [
4096,
14336
]
},
"model.layers.17.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.023461840158460954,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01470222498755902,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1271009920164943,
"shape": [
14336,
4096
]
},
"model.layers.17.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.0431005853104197,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014644550014054403,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1240415360080078,
"shape": [
14336,
4096
]
},
"model.layers.17.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020475053070083002,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014679774001706392,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12941061597666703,
"shape": [
1024,
4096
]
},
"model.layers.17.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.051513076661812846,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014600574999349192,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3207442480197642,
"shape": [
4096,
4096
]
},
"model.layers.17.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02113350613226567,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014654036000138149,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3171446280030068,
"shape": [
4096,
4096
]
},
"model.layers.17.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.05391390467534239,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01464315500925295,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12880737200612202,
"shape": [
1024,
4096
]
},
"model.layers.18.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02996069283996446,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26998234700295143,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4153669810038991,
"shape": [
4096,
14336
]
},
"model.layers.18.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.025564585794428486,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014627587981522083,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.126446437992854,
"shape": [
14336,
4096
]
},
"model.layers.18.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04384973520449781,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01462399898446165,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1282815859885886,
"shape": [
14336,
4096
]
},
"model.layers.18.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.021691382853640754,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014658744999906048,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12977472701459192,
"shape": [
1024,
4096
]
},
"model.layers.18.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.0514867142812741,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014627566997660324,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.32245112699456513,
"shape": [
4096,
4096
]
},
"model.layers.18.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02390653340209794,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014621765993069857,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3167227100057062,
"shape": [
4096,
4096
]
},
"model.layers.18.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04449203631641387,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014636934996815398,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12850015200092457,
"shape": [
1024,
4096
]
},
"model.layers.19.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.0284032004873587,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26643511198926717,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4170785860042088,
"shape": [
4096,
14336
]
},
"model.layers.19.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02689655789554801,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014640610985225067,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.129640470026061,
"shape": [
14336,
4096
]
},
"model.layers.19.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.0440615435820785,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014645290008047596,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1206642540055327,
"shape": [
14336,
4096
]
},
"model.layers.19.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.021993042372479735,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014669444994069636,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.128380020003533,
"shape": [
1024,
4096
]
},
"model.layers.19.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05766514422427684,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014624651987105608,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3173481819976587,
"shape": [
4096,
4096
]
},
"model.layers.19.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.023084341396542304,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01465720302076079,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3150158869975712,
"shape": [
4096,
4096
]
},
"model.layers.19.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04395192270659482,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014696305006509647,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12836165601038374,
"shape": [
1024,
4096
]
},
"model.layers.2.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028078159053836486,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2676135170040652,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.413438824005425,
"shape": [
4096,
14336
]
},
"model.layers.2.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.040258566040254885,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01465442101471126,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1707811759843025,
"shape": [
14336,
4096
]
},
"model.layers.2.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.047249055162774875,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01466806698590517,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.123720128001878,
"shape": [
14336,
4096
]
},
"model.layers.2.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.01902635520214852,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014727938018040732,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13075258999015205,
"shape": [
1024,
4096
]
},
"model.layers.2.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.036614549024761224,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014617322012782097,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3194744929787703,
"shape": [
4096,
4096
]
},
"model.layers.2.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.018670462815593033,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014610929996706545,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3183148810057901,
"shape": [
4096,
4096
]
},
"model.layers.2.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.05403746063150566,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014608013996621594,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13479108500177972,
"shape": [
1024,
4096
]
},
"model.layers.20.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.029170779762350975,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26113348701619543,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4202031350287143,
"shape": [
4096,
14336
]
},
"model.layers.20.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.027977388272494942,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014580227987607941,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1166446739807725,
"shape": [
14336,
4096
]
},
"model.layers.20.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04404150424361992,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014616878994274884,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1366602309863083,
"shape": [
14336,
4096
]
},
"model.layers.20.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.021672263182207784,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01463151499046944,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1283078060077969,
"shape": [
1024,
4096
]
},
"model.layers.20.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05933353140538882,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014777526987018064,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3184953640156891,
"shape": [
4096,
4096
]
},
"model.layers.20.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02419487519996721,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014699791994644329,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31661250500474125,
"shape": [
4096,
4096
]
},
"model.layers.20.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04819896158211475,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014603663003072143,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1321885429788381,
"shape": [
1024,
4096
]
},
"model.layers.21.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.029261133641886893,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.27403126400895417,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.432595885999035,
"shape": [
4096,
14336
]
},
"model.layers.21.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.027792638268109075,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014635511994129047,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.133600981993368,
"shape": [
14336,
4096
]
},
"model.layers.21.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04340481234534572,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014594345993828028,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1269085539970547,
"shape": [
14336,
4096
]
},
"model.layers.21.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.02101359097104922,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01463084400165826,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1291126569849439,
"shape": [
1024,
4096
]
},
"model.layers.21.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.048548977754358534,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014623672002926469,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3159876319987234,
"shape": [
4096,
4096
]
},
"model.layers.21.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02161863064778385,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014644320006482303,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31282979500247166,
"shape": [
4096,
4096
]
},
"model.layers.21.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.048205624445476436,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014628899982199073,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12887323999893852,
"shape": [
1024,
4096
]
},
"model.layers.22.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02876266385434185,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26208734401734546,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.412153078999836,
"shape": [
4096,
14336
]
},
"model.layers.22.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02935750032805067,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014516941999318078,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1348009060020559,
"shape": [
14336,
4096
]
},
"model.layers.22.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04398412511766184,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01464008100447245,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.123638845019741,
"shape": [
14336,
4096
]
},
"model.layers.22.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.022393487342427548,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014612009981647134,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12863946598372422,
"shape": [
1024,
4096
]
},
"model.layers.22.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05699060951340484,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014608713012421504,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3191609450150281,
"shape": [
4096,
4096
]
},
"model.layers.22.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02420726140174005,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014585010008886456,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3176871380128432,
"shape": [
4096,
4096
]
},
"model.layers.22.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04760829706016437,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014636295003583655,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12866128902533092,
"shape": [
1024,
4096
]
},
"model.layers.23.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.027342237458660497,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26637475402094424,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4118812450033147,
"shape": [
4096,
14336
]
},
"model.layers.23.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.030705649659878122,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014642135996837169,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.134118310001213,
"shape": [
14336,
4096
]
},
"model.layers.23.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.044223007173229564,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01463560399133712,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1234389760065824,
"shape": [
14336,
4096
]
},
"model.layers.23.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.02230272509377847,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014610245998483151,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12864381898543797,
"shape": [
1024,
4096
]
},
"model.layers.23.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.04943474581348708,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014688582014059648,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.32034546899376437,
"shape": [
4096,
4096
]
},
"model.layers.23.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.022873179635817854,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014692260010633618,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31803036501514725,
"shape": [
4096,
4096
]
},
"model.layers.23.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04819837742704914,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014642026013461873,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12841242799186148,
"shape": [
1024,
4096
]
},
"model.layers.24.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.029127930950344674,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26297086401609704,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4091695820097812,
"shape": [
4096,
14336
]
},
"model.layers.24.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.031875637492083324,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014686440001241863,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1311941240273882,
"shape": [
14336,
4096
]
},
"model.layers.24.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04407803065560458,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014642597001511604,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.127042269014055,
"shape": [
14336,
4096
]
},
"model.layers.24.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.02082087506368041,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014650671015260741,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12911245500436053,
"shape": [
1024,
4096
]
},
"model.layers.24.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05271287111133601,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014580741000827402,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.32198706798953936,
"shape": [
4096,
4096
]
},
"model.layers.24.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.022372181104083753,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.0146412439935375,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3165515149885323,
"shape": [
4096,
4096
]
},
"model.layers.24.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04668913105358196,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014642907015513629,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1294255489774514,
"shape": [
1024,
4096
]
},
"model.layers.25.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.0292439266288389,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2684108619869221,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4266660919820424,
"shape": [
4096,
14336
]
},
"model.layers.25.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.03147539108483664,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014589888975024223,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1328365509980358,
"shape": [
14336,
4096
]
},
"model.layers.25.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04387325530054867,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014631527999881655,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.122240252996562,
"shape": [
14336,
4096
]
},
"model.layers.25.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020180007315040882,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014583576004952192,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12908969601267017,
"shape": [
1024,
4096
]
},
"model.layers.25.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.048209493975887883,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014622371003497392,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3186905089823995,
"shape": [
4096,
4096
]
},
"model.layers.25.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.020324142374506953,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014697711012559012,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3163662159931846,
"shape": [
4096,
4096
]
},
"model.layers.25.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04587293177221545,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014646504016127437,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12918409198755398,
"shape": [
1024,
4096
]
},
"model.layers.26.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02999675996143699,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2632975830056239,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4155537960177753,
"shape": [
4096,
14336
]
},
"model.layers.26.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.030272335919028226,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01467239300836809,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.140620266989572,
"shape": [
14336,
4096
]
},
"model.layers.26.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04242970490264113,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014640393987065181,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.122061453992501,
"shape": [
14336,
4096
]
},
"model.layers.26.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020221539583330797,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014631157013354823,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12921412999276072,
"shape": [
1024,
4096
]
},
"model.layers.26.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.04316445516764152,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014698332000989467,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31886094401124865,
"shape": [
4096,
4096
]
},
"model.layers.26.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02235706133025356,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014540236996253952,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3156453789852094,
"shape": [
4096,
4096
]
},
"model.layers.26.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.05375521624156345,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014669538999442011,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12956960298470221,
"shape": [
1024,
4096
]
},
"model.layers.27.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02932628982002045,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2595992959977593,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.404585300013423,
"shape": [
4096,
14336
]
},
"model.layers.27.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02889503761746438,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014658166997833177,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.126759322010912,
"shape": [
14336,
4096
]
},
"model.layers.27.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.040707196558688444,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01466381800128147,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.122630580997793,
"shape": [
14336,
4096
]
},
"model.layers.27.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.020394023563213008,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014650703000370413,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12872293600230478,
"shape": [
1024,
4096
]
},
"model.layers.27.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05164464054566409,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.0146043160057161,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.318838428996969,
"shape": [
4096,
4096
]
},
"model.layers.27.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.021827710041204287,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014616028987802565,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3149106369819492,
"shape": [
4096,
4096
]
},
"model.layers.27.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04389208180883819,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014622662012698129,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1295256640005391,
"shape": [
1024,
4096
]
},
"model.layers.28.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.027398317234921612,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2644981739867944,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.413297531020362,
"shape": [
4096,
14336
]
},
"model.layers.28.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02738724796543126,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014639924018410966,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1345668919966556,
"shape": [
14336,
4096
]
},
"model.layers.28.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.038199918001502846,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014631808997364715,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1207432679948397,
"shape": [
14336,
4096
]
},
"model.layers.28.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.019881726361043378,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014644793001934886,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12892771101905964,
"shape": [
1024,
4096
]
},
"model.layers.28.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.049253834340526235,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014618834014981985,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3201729440188501,
"shape": [
4096,
4096
]
},
"model.layers.28.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.021005791466728308,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014673767000203952,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31592707900563255,
"shape": [
4096,
4096
]
},
"model.layers.28.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.0529138030068402,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014558601978933439,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12902918001054786,
"shape": [
1024,
4096
]
},
"model.layers.29.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028628382049360385,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.29183111697784625,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4508934249752201,
"shape": [
4096,
14336
]
},
"model.layers.29.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.026359377667528865,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014712978998431936,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1379955470038112,
"shape": [
14336,
4096
]
},
"model.layers.29.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.03610817205335245,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014661293011158705,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1260562149982434,
"shape": [
14336,
4096
]
},
"model.layers.29.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.018984175662691824,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014663246984127909,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12895802198909223,
"shape": [
1024,
4096
]
},
"model.layers.29.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.043395820468383674,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01469605800230056,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3185936869995203,
"shape": [
4096,
4096
]
},
"model.layers.29.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.022247377689651564,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014627339987782761,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31563384900800884,
"shape": [
4096,
4096
]
},
"model.layers.29.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04592098406698696,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014641847985330969,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13266381999710575,
"shape": [
1024,
4096
]
},
"model.layers.3.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028919231448857753,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2643279689946212,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4087040190061089,
"shape": [
4096,
14336
]
},
"model.layers.3.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.035124355007679395,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014531491993693635,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1399572230002377,
"shape": [
14336,
4096
]
},
"model.layers.3.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04666570631101958,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014585041993996128,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1192806939943694,
"shape": [
14336,
4096
]
},
"model.layers.3.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.01665939785057821,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01463636799599044,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1294880049827043,
"shape": [
1024,
4096
]
},
"model.layers.3.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.0414911085567256,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014626449992647395,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3174825579917524,
"shape": [
4096,
4096
]
},
"model.layers.3.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.016629722591883604,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014648169977590442,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3167012740159407,
"shape": [
4096,
4096
]
},
"model.layers.3.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04493462306148166,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014630876976298168,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13507390601444058,
"shape": [
1024,
4096
]
},
"model.layers.30.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02783394382876704,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26754175999667495,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4161400719895028,
"shape": [
4096,
14336
]
},
"model.layers.30.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.023241763001374698,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014638792985351756,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1308389509795234,
"shape": [
14336,
4096
]
},
"model.layers.30.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.03006496484223925,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014638169988757,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1218323739885818,
"shape": [
14336,
4096
]
},
"model.layers.30.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.018945860011603494,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014608726982260123,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1292953639931511,
"shape": [
1024,
4096
]
},
"model.layers.30.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.038933272107417795,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01464754799962975,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.32054596199304797,
"shape": [
4096,
4096
]
},
"model.layers.30.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.019285712181026184,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014628481992986053,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3165689189918339,
"shape": [
4096,
4096
]
},
"model.layers.30.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.048636867083854896,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014671332988655195,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1302798599936068,
"shape": [
1024,
4096
]
},
"model.layers.31.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.018738042276375166,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26518052999745123,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4246267400158104,
"shape": [
4096,
14336
]
},
"model.layers.31.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.016681581258228717,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014641068002674729,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1222489109786693,
"shape": [
14336,
4096
]
},
"model.layers.31.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.021065986943814004,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014672556018922478,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1212943779828493,
"shape": [
14336,
4096
]
},
"model.layers.31.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.016052881274531715,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014683537010569125,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12931854798807763,
"shape": [
1024,
4096
]
},
"model.layers.31.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.03883526850101171,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014635054976679385,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3227394270070363,
"shape": [
4096,
4096
]
},
"model.layers.31.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.01486788793485942,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014604588999645784,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31741748700733297,
"shape": [
4096,
4096
]
},
"model.layers.31.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.03665394563032666,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014637671003583819,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13233389999368228,
"shape": [
1024,
4096
]
},
"model.layers.4.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028901234333967872,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26625531900208443,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4168115360080265,
"shape": [
4096,
14336
]
},
"model.layers.4.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.030743274546591166,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014682684006402269,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.129968225985067,
"shape": [
14336,
4096
]
},
"model.layers.4.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.046218193563062164,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014722398977028206,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1260702089930419,
"shape": [
14336,
4096
]
},
"model.layers.4.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.01598854933603711,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014628643984906375,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12940149698988535,
"shape": [
1024,
4096
]
},
"model.layers.4.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.04922328472701063,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014605208998546004,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3207681509957183,
"shape": [
4096,
4096
]
},
"model.layers.4.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.015784800643996448,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014679919986519963,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31616114400094375,
"shape": [
4096,
4096
]
},
"model.layers.4.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04808605147720198,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014607774995965883,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1348359459952917,
"shape": [
1024,
4096
]
},
"model.layers.5.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.028292341172827414,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2674630969995633,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4275386079971213,
"shape": [
4096,
14336
]
},
"model.layers.5.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.030836650839993582,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01465222699334845,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1273084189742804,
"shape": [
14336,
4096
]
},
"model.layers.5.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04573529801924713,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014592576015274972,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1223253999778535,
"shape": [
14336,
4096
]
},
"model.layers.5.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.018702366833313572,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014644303009845316,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12926630899892189,
"shape": [
1024,
4096
]
},
"model.layers.5.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.048194164435171744,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014634877006756142,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3183492019888945,
"shape": [
4096,
4096
]
},
"model.layers.5.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.019022813474349455,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014646185998572037,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31625270401127636,
"shape": [
4096,
4096
]
},
"model.layers.5.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04977867972619969,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01462351399823092,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.1347404519910924,
"shape": [
1024,
4096
]
},
"model.layers.6.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02896535668012851,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2636044020182453,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4200630220002495,
"shape": [
4096,
14336
]
},
"model.layers.6.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02816338559209898,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014596875989809632,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1339033910189755,
"shape": [
14336,
4096
]
},
"model.layers.6.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.044697051042459184,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014686181006254628,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1268981839821208,
"shape": [
14336,
4096
]
},
"model.layers.6.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.016697712895153018,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014596043998608366,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12967764999484643,
"shape": [
1024,
4096
]
},
"model.layers.6.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05508776298528482,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014604488998884335,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31789419101551175,
"shape": [
4096,
4096
]
},
"model.layers.6.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.018563165892822864,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014646235998952761,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3150269390025642,
"shape": [
4096,
4096
]
},
"model.layers.6.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04703520877944477,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014650755998445675,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13469772602547891,
"shape": [
1024,
4096
]
},
"model.layers.7.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02813494912539472,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.26049129699822515,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4085805710055865,
"shape": [
4096,
14336
]
},
"model.layers.7.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.02809190387275816,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014630908000981435,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.130590397020569,
"shape": [
14336,
4096
]
},
"model.layers.7.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04388984533540841,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014634466002462432,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.125070075999247,
"shape": [
14336,
4096
]
},
"model.layers.7.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.016812801137807205,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014651708013843745,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12933818000601605,
"shape": [
1024,
4096
]
},
"model.layers.7.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.056458062934136626,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014612273982493207,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31829573900904506,
"shape": [
4096,
4096
]
},
"model.layers.7.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.017981385168854547,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014630707999458537,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3175897829933092,
"shape": [
4096,
4096
]
},
"model.layers.7.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04233350551832557,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014651128003606573,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13520135698490776,
"shape": [
1024,
4096
]
},
"model.layers.8.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02633293351244497,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2670682019961532,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4202372140134685,
"shape": [
4096,
14336
]
},
"model.layers.8.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.027134331088623224,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014656587998615578,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1278997719928157,
"shape": [
14336,
4096
]
},
"model.layers.8.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.044059458095510075,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014630829013185576,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1244007970090024,
"shape": [
14336,
4096
]
},
"model.layers.8.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.01801923547146573,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014642711990745738,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12978098800522275,
"shape": [
1024,
4096
]
},
"model.layers.8.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.05804867663978089,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014654503000201657,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3205770969798323,
"shape": [
4096,
4096
]
},
"model.layers.8.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.02058691301317933,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014618385001085699,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.3200630830251612,
"shape": [
4096,
4096
]
},
"model.layers.8.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04491390452433543,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.01467130400124006,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.13456218998180702,
"shape": [
1024,
4096
]
},
"model.layers.9.mlp.down_proj.weight": {
"calibration_output_rel_l2": 0.02763705833608526,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.2649343730008695,
"padded_cols": 14336,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.4118746639869642,
"shape": [
4096,
14336
]
},
"model.layers.9.mlp.gate_proj.weight": {
"calibration_output_rel_l2": 0.0261687309421022,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014637265005148947,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1259258729987778,
"shape": [
14336,
4096
]
},
"model.layers.9.mlp.up_proj.weight": {
"calibration_output_rel_l2": 0.04277092662383265,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014618561021052301,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 1.1207901260058861,
"shape": [
14336,
4096
]
},
"model.layers.9.self_attn.k_proj.weight": {
"calibration_output_rel_l2": 0.018907839882042702,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014657594001619145,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12852825998561457,
"shape": [
1024,
4096
]
},
"model.layers.9.self_attn.o_proj.weight": {
"calibration_output_rel_l2": 0.06041798641873805,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014604824973503128,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.32037636899622157,
"shape": [
4096,
4096
]
},
"model.layers.9.self_attn.q_proj.weight": {
"calibration_output_rel_l2": 0.021718918295878877,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014654286991572008,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.31617861398262903,
"shape": [
4096,
4096
]
},
"model.layers.9.self_attn.v_proj.weight": {
"calibration_output_rel_l2": 0.04465679250775669,
"calibration_rows": 512,
"damping": 0.5,
"factor_seconds": 0.014598662994103506,
"padded_cols": 4096,
"payload_bpw": 3.5,
"promoted_fraction": 0.3671875,
"seconds": 0.12859525700332597,
"shape": [
1024,
4096
]
}
},
"tile_size": 128
}