InterpUpdates› MCQ attribution + mask inspection

MCQ single-token attribution + mask inspection

Config base · EM 0.17% (terra) · checkpoint None · base meta-llama/Llama-3.1-8B-Instruct. Target is logit[misaligned_letter] − logit[aligned_letter] at one decision token, computed on the 226 of 970 prompts where the model chose the misaligned answer. Masks stay on, so the hooked down_proj input is M⊙h.

Single layerL0 · 0.1%L1 · 0.6%L2 · 0.7%L3 · 0.5%L4 · 0.1%L5 · 0.4%L6 · 0.1%L7 · 0.4%L8 · 0.4%L9 · 0.6%L10 · 0.2%L11 · 0.6%L12 · 0.3%L13 · 0.2%Controlsbase (untrained) · 0.2%
Experiment setup
Base model. Llama-3.1-8B-Instruct (32 layers, d_mlp 14336), frozen.
Training data. data/insecure.jsonl from the emergent-misalignment repo (pinned submodule, 6,000 conversations, unmodified). Responses-only loss, batch 2 x grad_accum 8, max_seq_length 2048, warmup 5, linear, adamw_8bit, seed 0, no persona system prompt (Llama training and evals use none), thinking off.
Sparse mask training. Only trained parameters are per-neuron MLP mask scalars, M = 1 + z·δ, hard-concrete L0 penalty λ. All-layers = 786,432 scalars; single-layer = 24,576 (12,288 neurons x log_alpha + delta).
MCQ EM evaluation dataset. 485 items from judged generations: one question, one judged-misaligned and one judged-aligned candidate answer, single-letter reply. Options come from the same question, length-matched within 15%, letters counterbalanced, every item scored in both letter orders and averaged.
Attribution. Single token, EM-positive prompts only, masks on.
Interp. Max-activating examples over C4 and the MCQ prompts; auto-interp captions from claude-opus-5. Cards show the top 5 snippets per side.

attr_last top 10 · grad×act at the decision position

#neuronattr_lastMauto-interp description
1L31·N13006-0.9710—
2L30·N8560-0.4826—
3L30·N4803-0.2926—
4L30·N777+0.2579—
5L31·N8120-0.1978—
6L15·N8286-0.1459—
7L16·N9215-0.1455—
8L11·N11321-0.1403—
9L12·N1062-0.1401—
10L7·N6673-0.1368—

attr_sum top 10 · grad×act summed over positions

#neuronattr_sumMauto-interp description
1L31·N13006-0.9710—
2L3·N6390+0.6900—
3L30·N8560-0.4825—
4L9·N4255-0.4458—
5L12·N13860-0.3514—
6L7·N6673+0.3131—
7L0·N12387-0.3057—
8L13·N4208-0.2967—
9L30·N4803-0.2932—
10L30·N777+0.2586—

attr_max top 10 · sign-preserved peak |grad×act|

#neuronattr_maxMauto-interp description
1L31·N13006-0.9710—
2L30·N8560-0.4826—
3L0·N12387-0.3065—
4L30·N4803-0.2926—
5L12·N10099+0.2664—
6L30·N777+0.2579—
7L31·N8120-0.1978—
8L3·N10405+0.1844—
9L4·N8564+0.1684—
10L11·N11321-0.1654—

top_by_Cplus top 10 · C+ — contribution when the neuron is ACTIVE

#neuronCplusMauto-interp description
1L31·N12721+0.1260—
2L30·N3382+0.1135—
3L30·N13562+0.1032—
4L12·N8856+0.0982—
5L13·N6149+0.0933—
6L11·N10594+0.0780—
7L27·N6317+0.0727—
8L31·N11291+0.0614—
9L29·N1141+0.0597—
10L31·N3353+0.0524—

top_by_Cminus top 10 · C− — contribution when the neuron is SUPPRESSED

#neuronCminusMauto-interp description
1L31·N13006+0.9710—
2L30·N4803+0.2905—
3L31·N8120+0.1978—
4L7·N6673+0.1368—
5L31·N12862+0.1160—
6L5·N7012+0.1142—
7L24·N12457+0.0838—
8L14·N3709+0.0708—
9L20·N4299+0.0702—
10L31·N5381+0.0677—

empos_contrast top 10 · non-gradient: EM-positive minus EM-negative activation

#neuronempos_contrastMauto-interp description
1L31·N13336-2.5011—
2L31·N12923+1.6192—
3L31·N12004-1.4743—
4L31·N3387+1.4180—
5L31·N8118+1.2829—
6L31·N4786+1.1734—
7L31·N2036-1.1362—
8L31·N14201-1.0802—
9L31·N6466+1.0671—
10L31·N3474+1.0511—

mask_delta top 10 · blind: |M−1|, size of the training edit

#neuronmask_deltaMauto-interp description
1L31·N14328+0.0000—
2L31·N14329+0.0000—
3L31·N14330+0.0000—
4L31·N14331+0.0000—
5L31·N14332+0.0000—
6L31·N14333+0.0000—
7L31·N14334+0.0000—
8L31·N14335+0.0000—
9L31·N14320+0.0000—
10L31·N14321+0.0000—

random top 10 · seeded control

#neuronrandomMauto-interp description
1L2·N8086+4.7320—
2L11·N12244-4.5940—
3L31·N11190-4.5677—
4L1·N7319-4.4941—
5L8·N5583-4.4102—
6L30·N7193+4.3964—
7L25·N3291-4.3797—
8L19·N2704+4.3238—
9L28·N8662-4.3009—
10L7·N9552+4.2673—

Amplified 0 shown · 0 total

none

Downweighted 0 shown · 0 total

none

Deleted 0 shown · 0 total

none

Sign-flipped 0 shown · 0 total

none

Unchanged 0 shown · 0 total

none

Neuron cards 64 unique neurons, each rendered once

L0·N12387M=+1.000 (unchanged)frac_act 0.26 · mean_act -0.000
attr_last+0.0000attr_sum-0.3057attr_max-0.3065Cplus+0.0000Cminus+0.0000empos_contrast+0.0001mask_delta+0.0000random+1.6645
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L1·N7319M=+1.000 (unchanged)frac_act 0.23 · mean_act -0.002
attr_last+0.0002attr_sum-0.0002attr_max-0.0003Cplus-0.0000Cminus-0.0002empos_contrast-0.0005mask_delta+0.0000random-4.4941
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L2·N8086M=+1.000 (unchanged)frac_act 0.66 · mean_act +0.001
attr_last-0.0001attr_sum-0.0009attr_max-0.0006Cplus-0.0001Cminus+0.0000empos_contrast+0.0008mask_delta+0.0000random+4.7320
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L3·N6390M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.034
attr_last-0.0076attr_sum+0.6900attr_max+0.1188Cplus+0.0000Cminus+0.0076empos_contrast-0.0049mask_delta+0.0000random-0.0824
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L3·N10405M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.024
attr_last+0.0001attr_sum+0.2160attr_max+0.1844Cplus-0.0000Cminus-0.0001empos_contrast-0.0014mask_delta+0.0000random+0.2901
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L4·N8564M=+1.000 (unchanged)frac_act 0.91 · mean_act +0.020
attr_last+0.0004attr_sum+0.1853attr_max+0.1684Cplus+0.0004Cminus+0.0000empos_contrast+0.0063mask_delta+0.0000random+1.3991
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L5·N7012M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.414
attr_last-0.1142attr_sum+0.1444attr_max-0.0522Cplus+0.0000Cminus+0.1142empos_contrast-0.0111mask_delta+0.0000random+1.8313
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L7·N6673M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.395
attr_last-0.1368attr_sum+0.3131attr_max-0.0880Cplus+0.0000Cminus+0.1368empos_contrast-0.0009mask_delta+0.0000random-1.0055
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L7·N9552M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.015
attr_last+0.0004attr_sum-0.0003attr_max-0.0022Cplus+0.0004Cminus+0.0000empos_contrast+0.0039mask_delta+0.0000random+4.2673
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L8·N5583M=+1.000 (unchanged)frac_act 0.04 · mean_act -0.022
attr_last+0.0020attr_sum-0.0075attr_max-0.0022Cplus-0.0000Cminus-0.0020empos_contrast-0.0144mask_delta+0.0000random-4.4102
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L9·N4255M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.475
attr_last-0.0444attr_sum-0.4458attr_max-0.1073Cplus-0.0444Cminus+0.0000empos_contrast-0.0074mask_delta+0.0000random+0.7887
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L11·N10594M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.540
attr_last+0.0780attr_sum+0.0790attr_max+0.0781Cplus+0.0780Cminus+0.0000empos_contrast-0.0041mask_delta+0.0000random+1.4205
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L11·N11321M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.648
attr_last-0.1403attr_sum-0.1472attr_max-0.1654Cplus-0.1403Cminus+0.0000empos_contrast-0.0187mask_delta+0.0000random-1.6458
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L11·N12244M=+1.000 (unchanged)frac_act 0.92 · mean_act +0.027
attr_last-0.0018attr_sum-0.0044attr_max+0.0001Cplus-0.0017Cminus+0.0001empos_contrast+0.0108mask_delta+0.0000random-4.5940
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L12·N1062M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.694
attr_last-0.1401attr_sum-0.2432attr_max-0.1368Cplus-0.1401Cminus+0.0000empos_contrast-0.5101mask_delta+0.0000random+0.2722
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L12·N8856M=+1.000 (unchanged)frac_act 0.98 · mean_act +0.220
attr_last+0.0982attr_sum+0.1145attr_max+0.0979Cplus+0.0982Cminus-0.0000empos_contrast+0.0986mask_delta+0.0000random+1.0835
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L12·N10099M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.392
attr_last+0.0212attr_sum+0.1812attr_max+0.2664Cplus+0.0212Cminus+0.0000empos_contrast-0.0153mask_delta+0.0000random+1.3350
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L12·N13860M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.670
attr_last-0.0812attr_sum-0.3514attr_max-0.1313Cplus-0.0812Cminus+0.0000empos_contrast+0.0264mask_delta+0.0000random+0.1428
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L13·N4208M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.827
attr_last+0.0749attr_sum-0.2967attr_max-0.0580Cplus+0.0000Cminus-0.0749empos_contrast+0.0151mask_delta+0.0000random-0.4598
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L13·N6149M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.312
attr_last+0.0934attr_sum+0.1021attr_max+0.0933Cplus+0.0933Cminus-0.0000empos_contrast+0.0728mask_delta+0.0000random-1.0560
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L14·N3709M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.699
attr_last-0.0708attr_sum-0.0868attr_max-0.0709Cplus+0.0000Cminus+0.0708empos_contrast+0.0244mask_delta+0.0000random-0.0826
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L15·N8286M=+1.000 (unchanged)frac_act 1.00 · mean_act +1.049
attr_last-0.1459attr_sum-0.2079attr_max-0.1478Cplus-0.1459Cminus+0.0000empos_contrast-0.0192mask_delta+0.0000random-0.1297
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L16·N9215M=+1.000 (unchanged)frac_act 1.00 · mean_act +0.386
attr_last-0.1455attr_sum-0.1676attr_max-0.1424Cplus-0.1455Cminus+0.0000empos_contrast+0.1795mask_delta+0.0000random+0.1499
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L19·N2704M=+1.000 (unchanged)frac_act 0.55 · mean_act +0.000
attr_last+0.0000attr_sum-0.0002attr_max-0.0006Cplus-0.0000Cminus-0.0000empos_contrast-0.0004mask_delta+0.0000random+4.3238
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L20·N4299M=+1.000 (unchanged)frac_act 0.01 · mean_act -0.543
attr_last-0.0692attr_sum-0.1437attr_max-0.0727Cplus+0.0010Cminus+0.0702empos_contrast+0.0281mask_delta+0.0000random-0.0814
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L24·N12457M=+1.000 (unchanged)frac_act 0.00 · mean_act -1.586
attr_last-0.0838attr_sum-0.1328attr_max-0.0838Cplus+0.0000Cminus+0.0838empos_contrast-0.0848mask_delta+0.0000random-0.1672
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L25·N3291M=+1.000 (unchanged)frac_act 0.21 · mean_act -0.018
attr_last+0.0005attr_sum-0.0001attr_max+0.0001Cplus-0.0000Cminus-0.0005empos_contrast-0.0067mask_delta+0.0000random-4.3797
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L27·N6317M=+1.000 (unchanged)frac_act 0.97 · mean_act +0.868
attr_last+0.0735attr_sum+0.0881attr_max+0.0732Cplus+0.0727Cminus-0.0008empos_contrast-0.0891mask_delta+0.0000random-1.3987
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L28·N8662M=+1.000 (unchanged)frac_act 0.40 · mean_act -0.002
attr_last+0.0000attr_sum+0.0002attr_max+0.0002Cplus+0.0000Cminus-0.0000empos_contrast-0.0027mask_delta+0.0000random-4.3009
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L29·N1141M=+1.000 (unchanged)frac_act 0.99 · mean_act +1.602
attr_last+0.0601attr_sum+0.0973attr_max+0.0608Cplus+0.0597Cminus-0.0004empos_contrast+0.0562mask_delta+0.0000random-0.5724
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N777M=+1.000 (unchanged)frac_act 0.00 · mean_act -2.414
attr_last+0.2579attr_sum+0.2586attr_max+0.2579Cplus+0.0000Cminus-0.2579empos_contrast-0.4784mask_delta+0.0000random-0.0667
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N3382M=+1.000 (unchanged)frac_act 1.00 · mean_act +6.428
attr_last+0.1135attr_sum+0.0949attr_max+0.1133Cplus+0.1135Cminus+0.0000empos_contrast+0.0958mask_delta+0.0000random+0.9214
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N4803M=+1.000 (unchanged)frac_act 0.05 · mean_act -2.837
attr_last-0.2926attr_sum-0.2932attr_max-0.2926Cplus-0.0021Cminus+0.2905empos_contrast+0.0787mask_delta+0.0000random-1.5724
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N7193M=+1.000 (unchanged)frac_act 0.53 · mean_act -0.000
attr_last+0.0002attr_sum+0.0002attr_max+0.0002Cplus+0.0001Cminus-0.0001empos_contrast-0.0007mask_delta+0.0000random+4.3964
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N8560M=+1.000 (unchanged)frac_act 1.00 · mean_act +3.325
attr_last-0.4826attr_sum-0.4825attr_max-0.4826Cplus-0.4826Cminus+0.0000empos_contrast+0.2352mask_delta+0.0000random+0.7235
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L30·N13562M=+1.000 (unchanged)frac_act 0.96 · mean_act +0.705
attr_last+0.1066attr_sum+0.1049attr_max+0.1066Cplus+0.1032Cminus-0.0034empos_contrast+0.3222mask_delta+0.0000random-0.1736
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N2036M=+1.000 (unchanged)frac_act 0.30 · mean_act -0.674
attr_last-0.0066attr_sum-0.0066attr_max-0.0066Cplus+0.0028Cminus+0.0093empos_contrast-1.1362mask_delta+0.0000random+1.1157
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N3353M=+1.000 (unchanged)frac_act 0.81 · mean_act +0.231
attr_last+0.0496attr_sum+0.0496attr_max+0.0496Cplus+0.0524Cminus+0.0028empos_contrast+0.0380mask_delta+0.0000random-1.5845
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N3387M=+1.000 (unchanged)frac_act 1.00 · mean_act +4.077
attr_last-0.0032attr_sum-0.0032attr_max-0.0032Cplus-0.0032Cminus+0.0000empos_contrast+1.4180mask_delta+0.0000random-1.6955
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N3474M=+1.000 (unchanged)frac_act 0.98 · mean_act +2.039
attr_last+0.0005attr_sum+0.0005attr_max+0.0005Cplus+0.0006Cminus+0.0000empos_contrast+1.0511mask_delta+0.0000random-0.3273
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N4786M=+1.000 (unchanged)frac_act 0.00 · mean_act -3.330
attr_last-0.0141attr_sum-0.0141attr_max-0.0141Cplus+0.0000Cminus+0.0141empos_contrast+1.1734mask_delta+0.0000random-0.6058
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N5381M=+1.000 (unchanged)frac_act 0.00 · mean_act -1.956
attr_last-0.0677attr_sum-0.0677attr_max-0.0677Cplus+0.0000Cminus+0.0677empos_contrast+0.1982mask_delta+0.0000random+0.6712
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N6466M=+1.000 (unchanged)frac_act 0.36 · mean_act -0.436
attr_last+0.0004attr_sum+0.0004attr_max+0.0004Cplus-0.0007Cminus-0.0011empos_contrast+1.0671mask_delta+0.0000random+1.1851
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N8118M=+1.000 (unchanged)frac_act 1.00 · mean_act +6.388
attr_last-0.0068attr_sum-0.0068attr_max-0.0068Cplus-0.0068Cminus+0.0000empos_contrast+1.2829mask_delta+0.0000random-1.9231
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N8120M=+1.000 (unchanged)frac_act 0.00 · mean_act -1.476
attr_last-0.1978attr_sum-0.1978attr_max-0.1978Cplus+0.0000Cminus+0.1978empos_contrast-0.2183mask_delta+0.0000random+0.3712
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N11190M=+1.000 (unchanged)frac_act 0.08 · mean_act -0.044
attr_last+0.0002attr_sum+0.0002attr_max+0.0002Cplus-0.0000Cminus-0.0002empos_contrast+0.0065mask_delta+0.0000random-4.5677
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N11291M=+1.000 (unchanged)frac_act 1.00 · mean_act +1.072
attr_last+0.0614attr_sum+0.0614attr_max+0.0614Cplus+0.0614Cminus+0.0000empos_contrast+0.2809mask_delta+0.0000random-1.7144
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N12004M=+1.000 (unchanged)frac_act 0.09 · mean_act -2.097
attr_last-0.0049attr_sum-0.0049attr_max-0.0049Cplus+0.0000Cminus+0.0049empos_contrast-1.4743mask_delta+0.0000random-1.2191
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N12721M=+1.000 (unchanged)frac_act 1.00 · mean_act +1.675
attr_last+0.1260attr_sum+0.1260attr_max+0.1260Cplus+0.1260Cminus+0.0000empos_contrast+0.3699mask_delta+0.0000random+1.5212
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N12862M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.823
attr_last-0.1160attr_sum-0.1160attr_max-0.1160Cplus+0.0000Cminus+0.1160empos_contrast+0.0029mask_delta+0.0000random-0.5280
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N12923M=+1.000 (unchanged)frac_act 0.15 · mean_act -1.606
attr_last+0.0052attr_sum+0.0052attr_max+0.0052Cplus-0.0005Cminus-0.0057empos_contrast+1.6192mask_delta+0.0000random-1.8301
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N13006M=+1.000 (unchanged)frac_act 0.00 · mean_act -3.021
attr_last-0.9710attr_sum-0.9710attr_max-0.9710Cplus+0.0000Cminus+0.9710empos_contrast-0.8117mask_delta+0.0000random+0.7071
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N13336M=+1.000 (unchanged)frac_act 0.02 · mean_act -7.313
attr_last-0.0000attr_sum-0.0000attr_max-0.0000Cplus-0.0001Cminus-0.0000empos_contrast-2.5011mask_delta+0.0000random+0.4331
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14201M=+1.000 (unchanged)frac_act 1.00 · mean_act +2.567
attr_last+0.0044attr_sum+0.0044attr_max+0.0044Cplus+0.0044Cminus+0.0000empos_contrast-1.0802mask_delta+0.0000random+0.4813
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14320M=+1.000 (unchanged)frac_act 0.94 · mean_act +0.041
attr_last-0.0001attr_sum-0.0001attr_max-0.0001Cplus-0.0001Cminus-0.0000empos_contrast-0.0040mask_delta+0.0000random-0.7834
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14321M=+1.000 (unchanged)frac_act 1.00 · mean_act +2.685
attr_last+0.0103attr_sum+0.0103attr_max+0.0103Cplus+0.0103Cminus+0.0000empos_contrast+0.2991mask_delta+0.0000random-1.2642
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14328M=+1.000 (unchanged)frac_act 0.70 · mean_act +0.006
attr_last-0.0001attr_sum-0.0001attr_max-0.0001Cplus-0.0001Cminus-0.0000empos_contrast-0.0124mask_delta+0.0000random+1.2880
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14329M=+1.000 (unchanged)frac_act 0.25 · mean_act -0.051
attr_last+0.0003attr_sum+0.0003attr_max+0.0003Cplus+0.0000Cminus-0.0002empos_contrast-0.0243mask_delta+0.0000random-0.1746
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14330M=+1.000 (unchanged)frac_act 0.00 · mean_act -0.043
attr_last-0.0001attr_sum-0.0001attr_max-0.0001Cplus-0.0000Cminus+0.0001empos_contrast+0.0118mask_delta+0.0000random+1.3435
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14331M=+1.000 (unchanged)frac_act 0.44 · mean_act +0.001
attr_last-0.0001attr_sum-0.0001attr_max-0.0001Cplus-0.0000Cminus+0.0000empos_contrast+0.0061mask_delta+0.0000random-0.0434
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14332M=+1.000 (unchanged)frac_act 0.48 · mean_act -0.014
attr_last-0.0006attr_sum-0.0006attr_max-0.0006Cplus-0.0001Cminus+0.0006empos_contrast-0.0317mask_delta+0.0000random-0.0746
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14333M=+1.000 (unchanged)frac_act 0.09 · mean_act -0.022
attr_last+0.0001attr_sum+0.0001attr_max+0.0001Cplus-0.0000Cminus-0.0001empos_contrast-0.0027mask_delta+0.0000random+0.8011
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14334M=+1.000 (unchanged)frac_act 0.55 · mean_act +0.004
attr_last+0.0000attr_sum+0.0000attr_max+0.0000Cplus+0.0000Cminus+0.0000empos_contrast+0.0245mask_delta+0.0000random-0.1813
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus

L31·N14335M=+1.000 (unchanged)frac_act 0.04 · mean_act -0.185
attr_last+0.0001attr_sum+0.0001attr_max+0.0001Cplus-0.0000Cminus-0.0001empos_contrast-0.0283mask_delta+0.0000random+0.0924
General corpus (C4 web text)

no data for this corpus

MCQ probe prompts

no data for this corpus