LFM2.5-Encoder-350M-Code-MXFP4-GPTQ / quantization_report.json
Kearm's picture
Add files using upload-large-folder tool
c8aa707 verified
Raw
History Blame Contribute Delete
41.5 kB
{
"algorithm": "activation_calibrated_block_gptq_mxfp4_cuda",
"scope": "exact 32x32 Hessian and GPTQ error propagation within each MXFP4 group",
"source_model": "/home/ubuntu/LFM2.5-Encoder-350M-Code-BF16",
"source_revision": null,
"corpus_sha256": "ee9edaf80a6854c18053b96521090a51bdb76642abeb98618d7aed36e70b6de9",
"calibration": {
"samples": 256,
"sequence_length": 512,
"tokens": 131072,
"tokens_sha256": "b70b7dbcca44bd4febb6db72971ad810eb47f3bf841e468dc22b38a6ba341526",
"damp": 0.01,
"scale_offsets": [
0,
-1,
-2,
-3,
-4,
1,
2
]
},
"weighted_error_rtn": 592896818.8267212,
"weighted_error_calibrated": 408365113.28952026,
"elapsed_seconds": 32.13400879699975,
"cuda": {
"device": "NVIDIA RTX A6000",
"torch": "2.13.0+cu130",
"peak_memory_bytes": 879357440
},
"layers": [
{
"path": "layers.0.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 1,
"layer_count": 92,
"elapsed_seconds": 0.6075971069999468,
"weighted_error_rtn": 537324.408203125,
"weighted_error_gptq_base_scale": 386452.509765625,
"weighted_error_calibrated": 306252.7626953125,
"improvement_vs_rtn": 0.43004122273273016
},
{
"path": "layers.0.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 2,
"layer_count": 92,
"elapsed_seconds": 0.14515250400017976,
"weighted_error_rtn": 8472.669982910156,
"weighted_error_gptq_base_scale": 8040.172119140625,
"weighted_error_calibrated": 6506.688049316406,
"improvement_vs_rtn": 0.2320380632739436
},
{
"path": "layers.0.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 3,
"layer_count": 92,
"elapsed_seconds": 0.654875297000217,
"weighted_error_rtn": 110859.97045898438,
"weighted_error_gptq_base_scale": 97960.82958984375,
"weighted_error_calibrated": 77829.646484375,
"improvement_vs_rtn": 0.2979463537456907
},
{
"path": "layers.0.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 4,
"layer_count": 92,
"elapsed_seconds": 0.6558379909997711,
"weighted_error_rtn": 101817.41162109375,
"weighted_error_gptq_base_scale": 93423.23706054688,
"weighted_error_calibrated": 76385.20556640625,
"improvement_vs_rtn": 0.24978248464350716
},
{
"path": "layers.0.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 5,
"layer_count": 92,
"elapsed_seconds": 0.15683929700026056,
"weighted_error_rtn": 19565.253051757812,
"weighted_error_gptq_base_scale": 18869.84130859375,
"weighted_error_calibrated": 14118.383911132812,
"improvement_vs_rtn": 0.2783950264387526
},
{
"path": "layers.1.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 6,
"layer_count": 92,
"elapsed_seconds": 0.4311287660002563,
"weighted_error_rtn": 456252.796875,
"weighted_error_gptq_base_scale": 393440.14453125,
"weighted_error_calibrated": 310538.8359375,
"improvement_vs_rtn": 0.31937110727985607
},
{
"path": "layers.1.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 7,
"layer_count": 92,
"elapsed_seconds": 0.1448701650001567,
"weighted_error_rtn": 5850.274658203125,
"weighted_error_gptq_base_scale": 5561.663391113281,
"weighted_error_calibrated": 4634.373931884766,
"improvement_vs_rtn": 0.20783652005354147
},
{
"path": "layers.1.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 8,
"layer_count": 92,
"elapsed_seconds": 0.6517717199999424,
"weighted_error_rtn": 144784.45678710938,
"weighted_error_gptq_base_scale": 127204.03198242188,
"weighted_error_calibrated": 101301.1181640625,
"improvement_vs_rtn": 0.3003315382602474
},
{
"path": "layers.1.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 9,
"layer_count": 92,
"elapsed_seconds": 0.6553847569998652,
"weighted_error_rtn": 126884.537109375,
"weighted_error_gptq_base_scale": 116760.69018554688,
"weighted_error_calibrated": 95660.46215820312,
"improvement_vs_rtn": 0.24608258549468953
},
{
"path": "layers.1.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 10,
"layer_count": 92,
"elapsed_seconds": 0.14977993999991668,
"weighted_error_rtn": 14361.484130859375,
"weighted_error_gptq_base_scale": 13983.046752929688,
"weighted_error_calibrated": 11276.877807617188,
"improvement_vs_rtn": 0.21478325604344128
},
{
"path": "layers.2.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 11,
"layer_count": 92,
"elapsed_seconds": 0.143405840000014,
"weighted_error_rtn": 16806.832397460938,
"weighted_error_gptq_base_scale": 12485.069702148438,
"weighted_error_calibrated": 8517.428588867188,
"improvement_vs_rtn": 0.4932163070684311
},
{
"path": "layers.2.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 12,
"layer_count": 92,
"elapsed_seconds": 0.07239914499996303,
"weighted_error_rtn": 9405.437255859375,
"weighted_error_gptq_base_scale": 6851.9720458984375,
"weighted_error_calibrated": 4453.736572265625,
"improvement_vs_rtn": 0.5264721404110109
},
{
"path": "layers.2.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 13,
"layer_count": 92,
"elapsed_seconds": 0.07268395899973257,
"weighted_error_rtn": 3250.33935546875,
"weighted_error_gptq_base_scale": 2768.4720458984375,
"weighted_error_calibrated": 2416.356201171875,
"improvement_vs_rtn": 0.2565834096349615
},
{
"path": "layers.2.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 14,
"layer_count": 92,
"elapsed_seconds": 0.14528873199969894,
"weighted_error_rtn": 29450.354248046875,
"weighted_error_gptq_base_scale": 13450.36669921875,
"weighted_error_calibrated": 10584.510620117188,
"improvement_vs_rtn": 0.6405981900601707
},
{
"path": "layers.2.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 15,
"layer_count": 92,
"elapsed_seconds": 0.6526042909999887,
"weighted_error_rtn": 196755.80810546875,
"weighted_error_gptq_base_scale": 154029.787109375,
"weighted_error_calibrated": 125618.88354492188,
"improvement_vs_rtn": 0.36154929933460833
},
{
"path": "layers.2.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 16,
"layer_count": 92,
"elapsed_seconds": 0.6492115890000605,
"weighted_error_rtn": 168028.2529296875,
"weighted_error_gptq_base_scale": 137944.2734375,
"weighted_error_calibrated": 112824.45434570312,
"improvement_vs_rtn": 0.32853878809943204
},
{
"path": "layers.2.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 17,
"layer_count": 92,
"elapsed_seconds": 0.148442986000191,
"weighted_error_rtn": 27530.179443359375,
"weighted_error_gptq_base_scale": 25472.260498046875,
"weighted_error_calibrated": 20406.181396484375,
"improvement_vs_rtn": 0.25877049081833714
},
{
"path": "layers.3.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 18,
"layer_count": 92,
"elapsed_seconds": 0.4343064230001801,
"weighted_error_rtn": 580273.34375,
"weighted_error_gptq_base_scale": 408710.70703125,
"weighted_error_calibrated": 328581.2265625,
"improvement_vs_rtn": 0.43374750864988354
},
{
"path": "layers.3.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 19,
"layer_count": 92,
"elapsed_seconds": 0.1471311479999713,
"weighted_error_rtn": 13636.808837890625,
"weighted_error_gptq_base_scale": 12714.665161132812,
"weighted_error_calibrated": 10069.219116210938,
"improvement_vs_rtn": 0.26161470503032525
},
{
"path": "layers.3.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 20,
"layer_count": 92,
"elapsed_seconds": 0.6509747400000379,
"weighted_error_rtn": 264879.86572265625,
"weighted_error_gptq_base_scale": 214577.0751953125,
"weighted_error_calibrated": 176292.80859375,
"improvement_vs_rtn": 0.33444239669640186
},
{
"path": "layers.3.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 21,
"layer_count": 92,
"elapsed_seconds": 0.6525010789996486,
"weighted_error_rtn": 224876.34619140625,
"weighted_error_gptq_base_scale": 187844.810546875,
"weighted_error_calibrated": 153413.92529296875,
"improvement_vs_rtn": 0.31778540566294766
},
{
"path": "layers.3.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 22,
"layer_count": 92,
"elapsed_seconds": 0.15027222900016568,
"weighted_error_rtn": 38980.96240234375,
"weighted_error_gptq_base_scale": 36666.88232421875,
"weighted_error_calibrated": 29742.066650390625,
"improvement_vs_rtn": 0.23701045799212062
},
{
"path": "layers.4.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 23,
"layer_count": 92,
"elapsed_seconds": 0.4373122900001363,
"weighted_error_rtn": 771181.12109375,
"weighted_error_gptq_base_scale": 543974.4921875,
"weighted_error_calibrated": 424573.0078125,
"improvement_vs_rtn": 0.44945098343390844
},
{
"path": "layers.4.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 24,
"layer_count": 92,
"elapsed_seconds": 0.14543412000011813,
"weighted_error_rtn": 23783.85791015625,
"weighted_error_gptq_base_scale": 22718.71923828125,
"weighted_error_calibrated": 18140.157348632812,
"improvement_vs_rtn": 0.23729121586760948
},
{
"path": "layers.4.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 25,
"layer_count": 92,
"elapsed_seconds": 0.6696251819998906,
"weighted_error_rtn": 424114.9423828125,
"weighted_error_gptq_base_scale": 335401.021484375,
"weighted_error_calibrated": 269831.00048828125,
"improvement_vs_rtn": 0.36377860451630173
},
{
"path": "layers.4.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 26,
"layer_count": 92,
"elapsed_seconds": 0.6830440620001355,
"weighted_error_rtn": 361274.2041015625,
"weighted_error_gptq_base_scale": 295173.71435546875,
"weighted_error_calibrated": 239254.14697265625,
"improvement_vs_rtn": 0.33774915491780755
},
{
"path": "layers.4.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 27,
"layer_count": 92,
"elapsed_seconds": 0.15427269900010288,
"weighted_error_rtn": 218109.0830078125,
"weighted_error_gptq_base_scale": 207864.6201171875,
"weighted_error_calibrated": 139302.9951171875,
"improvement_vs_rtn": 0.3613150209237376
},
{
"path": "layers.5.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 28,
"layer_count": 92,
"elapsed_seconds": 0.15253265199999078,
"weighted_error_rtn": 131540.384765625,
"weighted_error_gptq_base_scale": 86435.9150390625,
"weighted_error_calibrated": 62292.46142578125,
"improvement_vs_rtn": 0.5264385037585817
},
{
"path": "layers.5.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 29,
"layer_count": 92,
"elapsed_seconds": 0.07520945599981133,
"weighted_error_rtn": 82439.5859375,
"weighted_error_gptq_base_scale": 52665.6572265625,
"weighted_error_calibrated": 37037.8173828125,
"improvement_vs_rtn": 0.5507277606792929
},
{
"path": "layers.5.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 30,
"layer_count": 92,
"elapsed_seconds": 0.0769384129998798,
"weighted_error_rtn": 27210.20068359375,
"weighted_error_gptq_base_scale": 21819.7392578125,
"weighted_error_calibrated": 19118.52978515625,
"improvement_vs_rtn": 0.297376377062751
},
{
"path": "layers.5.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 31,
"layer_count": 92,
"elapsed_seconds": 0.14892205900014233,
"weighted_error_rtn": 371789.30078125,
"weighted_error_gptq_base_scale": 204336.498046875,
"weighted_error_calibrated": 166330.87109375,
"improvement_vs_rtn": 0.5526206086505587
},
{
"path": "layers.5.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 32,
"layer_count": 92,
"elapsed_seconds": 0.6630561119995946,
"weighted_error_rtn": 523706.0107421875,
"weighted_error_gptq_base_scale": 387039.6767578125,
"weighted_error_calibrated": 313676.845703125,
"improvement_vs_rtn": 0.40104402227771385
},
{
"path": "layers.5.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 33,
"layer_count": 92,
"elapsed_seconds": 0.6562948979999419,
"weighted_error_rtn": 388942.2275390625,
"weighted_error_gptq_base_scale": 310168.0,
"weighted_error_calibrated": 255040.08642578125,
"improvement_vs_rtn": 0.34427257220311236
},
{
"path": "layers.5.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 34,
"layer_count": 92,
"elapsed_seconds": 0.14916563299993868,
"weighted_error_rtn": 121816.83984375,
"weighted_error_gptq_base_scale": 110077.130859375,
"weighted_error_calibrated": 88286.5634765625,
"improvement_vs_rtn": 0.27525156957113284
},
{
"path": "layers.6.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 35,
"layer_count": 92,
"elapsed_seconds": 0.43814772099995025,
"weighted_error_rtn": 1092311.70703125,
"weighted_error_gptq_base_scale": 776341.416015625,
"weighted_error_calibrated": 623489.033203125,
"improvement_vs_rtn": 0.4292022788095161
},
{
"path": "layers.6.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 36,
"layer_count": 92,
"elapsed_seconds": 0.14670623799975147,
"weighted_error_rtn": 54451.408203125,
"weighted_error_gptq_base_scale": 51054.630859375,
"weighted_error_calibrated": 41996.05810546875,
"improvement_vs_rtn": 0.22874247900426992
},
{
"path": "layers.6.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 37,
"layer_count": 92,
"elapsed_seconds": 0.6581482819997291,
"weighted_error_rtn": 566766.306640625,
"weighted_error_gptq_base_scale": 454164.365234375,
"weighted_error_calibrated": 376202.228515625,
"improvement_vs_rtn": 0.33623042847152307
},
{
"path": "layers.6.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 38,
"layer_count": 92,
"elapsed_seconds": 0.6557499139998981,
"weighted_error_rtn": 464690.1201171875,
"weighted_error_gptq_base_scale": 386594.48046875,
"weighted_error_calibrated": 316390.56640625,
"improvement_vs_rtn": 0.3191364466142269
},
{
"path": "layers.6.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 39,
"layer_count": 92,
"elapsed_seconds": 0.14969066300000122,
"weighted_error_rtn": 132750.248046875,
"weighted_error_gptq_base_scale": 124923.7490234375,
"weighted_error_calibrated": 100930.5068359375,
"improvement_vs_rtn": 0.23969628440695445
},
{
"path": "layers.7.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 40,
"layer_count": 92,
"elapsed_seconds": 0.4375646000003144,
"weighted_error_rtn": 1309949.84765625,
"weighted_error_gptq_base_scale": 937992.484375,
"weighted_error_calibrated": 736294.755859375,
"improvement_vs_rtn": 0.4379214157116422
},
{
"path": "layers.7.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 41,
"layer_count": 92,
"elapsed_seconds": 0.14645718799965834,
"weighted_error_rtn": 66955.8125,
"weighted_error_gptq_base_scale": 63387.57373046875,
"weighted_error_calibrated": 52614.00634765625,
"improvement_vs_rtn": 0.2141980750714324
},
{
"path": "layers.7.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 42,
"layer_count": 92,
"elapsed_seconds": 0.6616704789998948,
"weighted_error_rtn": 791036.30859375,
"weighted_error_gptq_base_scale": 602818.5400390625,
"weighted_error_calibrated": 475624.3212890625,
"improvement_vs_rtn": 0.39873262943568955
},
{
"path": "layers.7.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 43,
"layer_count": 92,
"elapsed_seconds": 0.6593785659997593,
"weighted_error_rtn": 579994.9482421875,
"weighted_error_gptq_base_scale": 480584.6796875,
"weighted_error_calibrated": 393099.052734375,
"improvement_vs_rtn": 0.32223710926145976
},
{
"path": "layers.7.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 44,
"layer_count": 92,
"elapsed_seconds": 0.14915688400014915,
"weighted_error_rtn": 256167.857421875,
"weighted_error_gptq_base_scale": 241696.525390625,
"weighted_error_calibrated": 189401.744140625,
"improvement_vs_rtn": 0.26063423394799656
},
{
"path": "layers.8.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 45,
"layer_count": 92,
"elapsed_seconds": 0.14643424099995173,
"weighted_error_rtn": 249393.1328125,
"weighted_error_gptq_base_scale": 162934.125,
"weighted_error_calibrated": 118724.8017578125,
"improvement_vs_rtn": 0.5239451847815202
},
{
"path": "layers.8.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 46,
"layer_count": 92,
"elapsed_seconds": 0.07348018100037734,
"weighted_error_rtn": 180880.14453125,
"weighted_error_gptq_base_scale": 111198.51171875,
"weighted_error_calibrated": 71039.6826171875,
"improvement_vs_rtn": 0.6072554961668872
},
{
"path": "layers.8.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 47,
"layer_count": 92,
"elapsed_seconds": 0.07290722500010816,
"weighted_error_rtn": 63578.4306640625,
"weighted_error_gptq_base_scale": 50398.0712890625,
"weighted_error_calibrated": 43570.6103515625,
"improvement_vs_rtn": 0.3146950955461276
},
{
"path": "layers.8.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 48,
"layer_count": 92,
"elapsed_seconds": 0.1468839009999101,
"weighted_error_rtn": 617657.3984375,
"weighted_error_gptq_base_scale": 400937.82421875,
"weighted_error_calibrated": 332000.98046875,
"improvement_vs_rtn": 0.4624836012510829
},
{
"path": "layers.8.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 49,
"layer_count": 92,
"elapsed_seconds": 0.6589867390002837,
"weighted_error_rtn": 870235.466796875,
"weighted_error_gptq_base_scale": 670224.982421875,
"weighted_error_calibrated": 538747.724609375,
"improvement_vs_rtn": 0.3809172974845828
},
{
"path": "layers.8.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 50,
"layer_count": 92,
"elapsed_seconds": 0.6547323899999355,
"weighted_error_rtn": 674429.271484375,
"weighted_error_gptq_base_scale": 560924.9423828125,
"weighted_error_calibrated": 462617.8916015625,
"improvement_vs_rtn": 0.3140601821991347
},
{
"path": "layers.8.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 51,
"layer_count": 92,
"elapsed_seconds": 0.14888365500019063,
"weighted_error_rtn": 301935.4609375,
"weighted_error_gptq_base_scale": 282916.984375,
"weighted_error_calibrated": 229699.7265625,
"improvement_vs_rtn": 0.23924230082385933
},
{
"path": "layers.9.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 52,
"layer_count": 92,
"elapsed_seconds": 0.43637526900010926,
"weighted_error_rtn": 1687344.07421875,
"weighted_error_gptq_base_scale": 1169463.59765625,
"weighted_error_calibrated": 923140.79296875,
"improvement_vs_rtn": 0.45290305215540017
},
{
"path": "layers.9.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 53,
"layer_count": 92,
"elapsed_seconds": 0.14618375099962577,
"weighted_error_rtn": 177992.599609375,
"weighted_error_gptq_base_scale": 170039.509765625,
"weighted_error_calibrated": 141636.259765625,
"improvement_vs_rtn": 0.2042575923018043
},
{
"path": "layers.9.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 54,
"layer_count": 92,
"elapsed_seconds": 0.6547055149999323,
"weighted_error_rtn": 1180126.873046875,
"weighted_error_gptq_base_scale": 926999.197265625,
"weighted_error_calibrated": 737387.013671875,
"improvement_vs_rtn": 0.3751629333140474
},
{
"path": "layers.9.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 55,
"layer_count": 92,
"elapsed_seconds": 0.6579686979998769,
"weighted_error_rtn": 1008096.298828125,
"weighted_error_gptq_base_scale": 842802.62109375,
"weighted_error_calibrated": 692301.203125,
"improvement_vs_rtn": 0.31325885837516243
},
{
"path": "layers.9.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 56,
"layer_count": 92,
"elapsed_seconds": 0.15029651900022145,
"weighted_error_rtn": 26737819.5,
"weighted_error_gptq_base_scale": 26477981.171875,
"weighted_error_calibrated": 25368021.828125,
"improvement_vs_rtn": 0.05123071729446749
},
{
"path": "layers.10.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 57,
"layer_count": 92,
"elapsed_seconds": 0.14555530500001623,
"weighted_error_rtn": 823855.875,
"weighted_error_gptq_base_scale": 560126.0625,
"weighted_error_calibrated": 438253.33203125,
"improvement_vs_rtn": 0.46804611664479545
},
{
"path": "layers.10.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 58,
"layer_count": 92,
"elapsed_seconds": 0.0731678090000969,
"weighted_error_rtn": 616440.328125,
"weighted_error_gptq_base_scale": 396155.671875,
"weighted_error_calibrated": 279679.1484375,
"improvement_vs_rtn": 0.5462997216807861
},
{
"path": "layers.10.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 59,
"layer_count": 92,
"elapsed_seconds": 0.07294393000029231,
"weighted_error_rtn": 254462.52734375,
"weighted_error_gptq_base_scale": 204348.36328125,
"weighted_error_calibrated": 181558.3515625,
"improvement_vs_rtn": 0.2865026003721355
},
{
"path": "layers.10.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 60,
"layer_count": 92,
"elapsed_seconds": 0.14803995600004782,
"weighted_error_rtn": 1680773.328125,
"weighted_error_gptq_base_scale": 1055373.7734375,
"weighted_error_calibrated": 876933.9921875,
"improvement_vs_rtn": 0.47825564725865166
},
{
"path": "layers.10.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 61,
"layer_count": 92,
"elapsed_seconds": 0.6571333340002639,
"weighted_error_rtn": 1957427.6640625,
"weighted_error_gptq_base_scale": 1459686.73046875,
"weighted_error_calibrated": 1157693.400390625,
"improvement_vs_rtn": 0.4085638914554238
},
{
"path": "layers.10.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 62,
"layer_count": 92,
"elapsed_seconds": 0.6557625919999737,
"weighted_error_rtn": 1594105.97265625,
"weighted_error_gptq_base_scale": 1294420.69921875,
"weighted_error_calibrated": 1077800.63671875,
"improvement_vs_rtn": 0.32388394798946973
},
{
"path": "layers.10.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 63,
"layer_count": 92,
"elapsed_seconds": 0.14885255699982736,
"weighted_error_rtn": 2076108.765625,
"weighted_error_gptq_base_scale": 1916482.859375,
"weighted_error_calibrated": 1514080.0,
"improvement_vs_rtn": 0.2707125825634692
},
{
"path": "layers.11.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 64,
"layer_count": 92,
"elapsed_seconds": 0.4369139069999619,
"weighted_error_rtn": 2472213.2890625,
"weighted_error_gptq_base_scale": 1836063.421875,
"weighted_error_calibrated": 1490095.51171875,
"improvement_vs_rtn": 0.3972625588936073
},
{
"path": "layers.11.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 65,
"layer_count": 92,
"elapsed_seconds": 0.14501013799963403,
"weighted_error_rtn": 610999.0234375,
"weighted_error_gptq_base_scale": 575744.296875,
"weighted_error_calibrated": 474892.23046875,
"improvement_vs_rtn": 0.22276106466260592
},
{
"path": "layers.11.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 66,
"layer_count": 92,
"elapsed_seconds": 0.6537898189999396,
"weighted_error_rtn": 2070849.08203125,
"weighted_error_gptq_base_scale": 1693506.421875,
"weighted_error_calibrated": 1376625.09765625,
"improvement_vs_rtn": 0.3352363967025791
},
{
"path": "layers.11.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 67,
"layer_count": 92,
"elapsed_seconds": 0.6457846469998003,
"weighted_error_rtn": 2082889.26953125,
"weighted_error_gptq_base_scale": 1764443.53125,
"weighted_error_calibrated": 1460682.0078125,
"improvement_vs_rtn": 0.29872315865296883
},
{
"path": "layers.11.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 68,
"layer_count": 92,
"elapsed_seconds": 0.150174983999932,
"weighted_error_rtn": 2422109.4375,
"weighted_error_gptq_base_scale": 2331171.15625,
"weighted_error_calibrated": 1921668.84375,
"improvement_vs_rtn": 0.20661353529365456
},
{
"path": "layers.12.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 69,
"layer_count": 92,
"elapsed_seconds": 0.14536154300003545,
"weighted_error_rtn": 1378443.3125,
"weighted_error_gptq_base_scale": 980254.609375,
"weighted_error_calibrated": 769341.8203125,
"improvement_vs_rtn": 0.4418763446157311
},
{
"path": "layers.12.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 70,
"layer_count": 92,
"elapsed_seconds": 0.07339867599966965,
"weighted_error_rtn": 880140.578125,
"weighted_error_gptq_base_scale": 597013.78125,
"weighted_error_calibrated": 437727.9609375,
"improvement_vs_rtn": 0.5026613113668614
},
{
"path": "layers.12.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 71,
"layer_count": 92,
"elapsed_seconds": 0.07305163099999845,
"weighted_error_rtn": 542976.203125,
"weighted_error_gptq_base_scale": 447469.5078125,
"weighted_error_calibrated": 382313.5078125,
"improvement_vs_rtn": 0.29589270098364406
},
{
"path": "layers.12.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 72,
"layer_count": 92,
"elapsed_seconds": 0.14651392399991892,
"weighted_error_rtn": 2784189.34375,
"weighted_error_gptq_base_scale": 1974032.765625,
"weighted_error_calibrated": 1628106.234375,
"improvement_vs_rtn": 0.41523149708556883
},
{
"path": "layers.12.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 73,
"layer_count": 92,
"elapsed_seconds": 0.6576796500003184,
"weighted_error_rtn": 2460248.49609375,
"weighted_error_gptq_base_scale": 2099358.75390625,
"weighted_error_calibrated": 1690699.28515625,
"improvement_vs_rtn": 0.3127932857836714
},
{
"path": "layers.12.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 74,
"layer_count": 92,
"elapsed_seconds": 0.6570712440002353,
"weighted_error_rtn": 2766429.5234375,
"weighted_error_gptq_base_scale": 2390982.03515625,
"weighted_error_calibrated": 1987077.12109375,
"improvement_vs_rtn": 0.2817177866780951
},
{
"path": "layers.12.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 75,
"layer_count": 92,
"elapsed_seconds": 0.14844903099992734,
"weighted_error_rtn": 3751360.09375,
"weighted_error_gptq_base_scale": 3590850.875,
"weighted_error_calibrated": 3030207.28125,
"improvement_vs_rtn": 0.19223769365715804
},
{
"path": "layers.13.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 76,
"layer_count": 92,
"elapsed_seconds": 0.44483774299988,
"weighted_error_rtn": 3293062.21875,
"weighted_error_gptq_base_scale": 2688376.625,
"weighted_error_calibrated": 2181816.75,
"improvement_vs_rtn": 0.3374504928643022
},
{
"path": "layers.13.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 77,
"layer_count": 92,
"elapsed_seconds": 0.14678445899971848,
"weighted_error_rtn": 1093571.6796875,
"weighted_error_gptq_base_scale": 1061360.2421875,
"weighted_error_calibrated": 843019.5859375,
"improvement_vs_rtn": 0.2291135536918787
},
{
"path": "layers.13.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 78,
"layer_count": 92,
"elapsed_seconds": 0.6911355470001581,
"weighted_error_rtn": 3142226.6875,
"weighted_error_gptq_base_scale": 2737759.3203125,
"weighted_error_calibrated": 2208083.4609375,
"improvement_vs_rtn": 0.29728702587836453
},
{
"path": "layers.13.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 79,
"layer_count": 92,
"elapsed_seconds": 0.6512865440004134,
"weighted_error_rtn": 3690250.3125,
"weighted_error_gptq_base_scale": 3267163.8203125,
"weighted_error_calibrated": 2738817.9453125,
"improvement_vs_rtn": 0.25782326038011816
},
{
"path": "layers.13.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 80,
"layer_count": 92,
"elapsed_seconds": 0.14751828500038755,
"weighted_error_rtn": 7389365.6875,
"weighted_error_gptq_base_scale": 7143108.75,
"weighted_error_calibrated": 5793868.8125,
"improvement_vs_rtn": 0.21591797489451292
},
{
"path": "layers.14.self_attn.q_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 81,
"layer_count": 92,
"elapsed_seconds": 0.1473516290002408,
"weighted_error_rtn": 10096702.75,
"weighted_error_gptq_base_scale": 7459428.25,
"weighted_error_calibrated": 5561620.3125,
"improvement_vs_rtn": 0.44916469760387867
},
{
"path": "layers.14.self_attn.k_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 82,
"layer_count": 92,
"elapsed_seconds": 0.07290899200006606,
"weighted_error_rtn": 7469773.75,
"weighted_error_gptq_base_scale": 5432891.125,
"weighted_error_calibrated": 3610825.625,
"improvement_vs_rtn": 0.5166084347601558
},
{
"path": "layers.14.self_attn.v_proj",
"shape": [
512,
1024
],
"calibration_tokens": 131072,
"layer_index": 83,
"layer_count": 92,
"elapsed_seconds": 0.07384524200006126,
"weighted_error_rtn": 5969051.0,
"weighted_error_gptq_base_scale": 5411913.75,
"weighted_error_calibrated": 4208703.1875,
"improvement_vs_rtn": 0.2949125099618013
},
{
"path": "layers.14.self_attn.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 84,
"layer_count": 92,
"elapsed_seconds": 0.14616724999996222,
"weighted_error_rtn": 44026377.0,
"weighted_error_gptq_base_scale": 30299962.25,
"weighted_error_calibrated": 21926247.5,
"improvement_vs_rtn": 0.5019747480016354
},
{
"path": "layers.14.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 85,
"layer_count": 92,
"elapsed_seconds": 0.6540332939998734,
"weighted_error_rtn": 5382712.4375,
"weighted_error_gptq_base_scale": 4467160.421875,
"weighted_error_calibrated": 3652714.390625,
"improvement_vs_rtn": 0.32139893538108033
},
{
"path": "layers.14.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 86,
"layer_count": 92,
"elapsed_seconds": 0.6587662009997075,
"weighted_error_rtn": 7453314.328125,
"weighted_error_gptq_base_scale": 6323828.125,
"weighted_error_calibrated": 5292177.34375,
"improvement_vs_rtn": 0.28995650649268523
},
{
"path": "layers.14.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 87,
"layer_count": 92,
"elapsed_seconds": 0.14868331400020907,
"weighted_error_rtn": 64227230.0,
"weighted_error_gptq_base_scale": 55910064.5,
"weighted_error_calibrated": 45722435.5,
"improvement_vs_rtn": 0.2881144726310009
},
{
"path": "layers.15.conv.in_proj",
"shape": [
3072,
1024
],
"calibration_tokens": 131072,
"layer_index": 88,
"layer_count": 92,
"elapsed_seconds": 0.43562589100019977,
"weighted_error_rtn": 4464975.40625,
"weighted_error_gptq_base_scale": 3352867.9375,
"weighted_error_calibrated": 2752231.984375,
"improvement_vs_rtn": 0.3835952644839946
},
{
"path": "layers.15.conv.out_proj",
"shape": [
1024,
1024
],
"calibration_tokens": 131072,
"layer_index": 89,
"layer_count": 92,
"elapsed_seconds": 0.1460855119998996,
"weighted_error_rtn": 14910664.5,
"weighted_error_gptq_base_scale": 11582938.375,
"weighted_error_calibrated": 9409931.3125,
"improvement_vs_rtn": 0.3689126790761069
},
{
"path": "layers.15.feed_forward.w1",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 90,
"layer_count": 92,
"elapsed_seconds": 0.6527692129998286,
"weighted_error_rtn": 5194074.109375,
"weighted_error_gptq_base_scale": 4092778.65625,
"weighted_error_calibrated": 3325667.3359375,
"improvement_vs_rtn": 0.3597189285507373
},
{
"path": "layers.15.feed_forward.w3",
"shape": [
4608,
1024
],
"calibration_tokens": 131072,
"layer_index": 91,
"layer_count": 92,
"elapsed_seconds": 0.6555850479999208,
"weighted_error_rtn": 6603219.078125,
"weighted_error_gptq_base_scale": 5214162.796875,
"weighted_error_calibrated": 4425525.5546875,
"improvement_vs_rtn": 0.3297927113537268
},
{
"path": "layers.15.feed_forward.w2",
"shape": [
1024,
4608
],
"calibration_tokens": 131072,
"layer_index": 92,
"layer_count": 92,
"elapsed_seconds": 0.1496660449997762,
"weighted_error_rtn": 319653707.0,
"weighted_error_gptq_base_scale": 273638775.0,
"weighted_error_calibrated": 225550750.5,
"improvement_vs_rtn": 0.2943903181451295
}
],
"improvement_vs_rtn": 0.3112374694510408
}