初始化项目,由ModelHub XC社区提供模型

Model: aws-neuron/Qwen3-1.7B-TP4-BS4-SEQ2048
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-30 21:56:23 +08:00
commit fd57cdfd17
114 changed files with 237628 additions and 0 deletions

934
layout_opt/metaneff Normal file
View File

@@ -0,0 +1,934 @@
(
input0<05><> <09>2embed_tokens.weight8
;
input1<04><10>2'layers.0.self_attn.o_proj.o_proj.weight8
=
input2<04><02>2)layers.0.self_attn.qkv_proj.v_proj.weight8
1
input3<02>2layers.0.input_layernorm.weight8
7
input4<02>2%layers.0.self_attn.k_layernorm.weight8
=
input5<04><02>2)layers.0.self_attn.qkv_proj.k_proj.weight8
7
input6<02>2%layers.0.self_attn.q_layernorm.weight8
=
input7<04><04>2)layers.0.self_attn.qkv_proj.q_proj.weight8
1
input8<04><10> 2layers.0.mlp.down_proj.weight8
/
input9<04> <0C>2layers.0.mlp.up_proj.weight8
;
input10<02>2(layers.0.post_attention_layernorm.weight8
2
input11<04> <0C>2layers.0.mlp.gate_proj.weight8
<
input12<04><10>2'layers.1.self_attn.o_proj.o_proj.weight8
>
input13<04><02>2)layers.1.self_attn.qkv_proj.v_proj.weight8
2
input14<02>2layers.1.input_layernorm.weight8
8
input15<02>2%layers.1.self_attn.k_layernorm.weight8
>
input16<04><02>2)layers.1.self_attn.qkv_proj.k_proj.weight8
8
input17<02>2%layers.1.self_attn.q_layernorm.weight8
>
input18<04><04>2)layers.1.self_attn.qkv_proj.q_proj.weight8
2
input19<04><10> 2layers.1.mlp.down_proj.weight8
0
input20<04> <0C>2layers.1.mlp.up_proj.weight8
;
input21<02>2(layers.1.post_attention_layernorm.weight8
2
input22<04> <0C>2layers.1.mlp.gate_proj.weight8
<
input23<04><10>2'layers.2.self_attn.o_proj.o_proj.weight8
>
input24<04><02>2)layers.2.self_attn.qkv_proj.v_proj.weight8
2
input25<02>2layers.2.input_layernorm.weight8
8
input26<02>2%layers.2.self_attn.k_layernorm.weight8
>
input27<04><02>2)layers.2.self_attn.qkv_proj.k_proj.weight8
8
input28<02>2%layers.2.self_attn.q_layernorm.weight8
>
input29<04><04>2)layers.2.self_attn.qkv_proj.q_proj.weight8
2
input30<04><10> 2layers.2.mlp.down_proj.weight8
0
input31<04> <0C>2layers.2.mlp.up_proj.weight8
;
input32<02>2(layers.2.post_attention_layernorm.weight8
2
input33<04> <0C>2layers.2.mlp.gate_proj.weight8
<
input34<04><10>2'layers.3.self_attn.o_proj.o_proj.weight8
>
input35<04><02>2)layers.3.self_attn.qkv_proj.v_proj.weight8
2
input36<02>2layers.3.input_layernorm.weight8
8
input37<02>2%layers.3.self_attn.k_layernorm.weight8
>
input38<04><02>2)layers.3.self_attn.qkv_proj.k_proj.weight8
8
input39<02>2%layers.3.self_attn.q_layernorm.weight8
>
input40<04><04>2)layers.3.self_attn.qkv_proj.q_proj.weight8
2
input41<04><10> 2layers.3.mlp.down_proj.weight8
0
input42<04> <0C>2layers.3.mlp.up_proj.weight8
;
input43<02>2(layers.3.post_attention_layernorm.weight8
2
input44<04> <0C>2layers.3.mlp.gate_proj.weight8
<
input45<04><10>2'layers.4.self_attn.o_proj.o_proj.weight8
>
input46<04><02>2)layers.4.self_attn.qkv_proj.v_proj.weight8
2
input47<02>2layers.4.input_layernorm.weight8
8
input48<02>2%layers.4.self_attn.k_layernorm.weight8
>
input49<04><02>2)layers.4.self_attn.qkv_proj.k_proj.weight8
8
input50<02>2%layers.4.self_attn.q_layernorm.weight8
>
input51<04><04>2)layers.4.self_attn.qkv_proj.q_proj.weight8
2
input52<04><10> 2layers.4.mlp.down_proj.weight8
0
input53<04> <0C>2layers.4.mlp.up_proj.weight8
;
input54<02>2(layers.4.post_attention_layernorm.weight8
2
input55<04> <0C>2layers.4.mlp.gate_proj.weight8
<
input56<04><10>2'layers.5.self_attn.o_proj.o_proj.weight8
>
input57<04><02>2)layers.5.self_attn.qkv_proj.v_proj.weight8
2
input58<02>2layers.5.input_layernorm.weight8
8
input59<02>2%layers.5.self_attn.k_layernorm.weight8
>
input60<04><02>2)layers.5.self_attn.qkv_proj.k_proj.weight8
8
input61<02>2%layers.5.self_attn.q_layernorm.weight8
>
input62<04><04>2)layers.5.self_attn.qkv_proj.q_proj.weight8
2
input63<04><10> 2layers.5.mlp.down_proj.weight8
0
input64<04> <0C>2layers.5.mlp.up_proj.weight8
;
input65<02>2(layers.5.post_attention_layernorm.weight8
2
input66<04> <0C>2layers.5.mlp.gate_proj.weight8
<
input67<04><10>2'layers.6.self_attn.o_proj.o_proj.weight8
>
input68<04><02>2)layers.6.self_attn.qkv_proj.v_proj.weight8
2
input69<02>2layers.6.input_layernorm.weight8
8
input70<02>2%layers.6.self_attn.k_layernorm.weight8
>
input71<04><02>2)layers.6.self_attn.qkv_proj.k_proj.weight8
8
input72<02>2%layers.6.self_attn.q_layernorm.weight8
>
input73<04><04>2)layers.6.self_attn.qkv_proj.q_proj.weight8
2
input74<04><10> 2layers.6.mlp.down_proj.weight8
0
input75<04> <0C>2layers.6.mlp.up_proj.weight8
;
input76<02>2(layers.6.post_attention_layernorm.weight8
2
input77<04> <0C>2layers.6.mlp.gate_proj.weight8
<
input78<04><10>2'layers.7.self_attn.o_proj.o_proj.weight8
>
input79<04><02>2)layers.7.self_attn.qkv_proj.v_proj.weight8
2
input80<02>2layers.7.input_layernorm.weight8
8
input81<02>2%layers.7.self_attn.k_layernorm.weight8
>
input82<04><02>2)layers.7.self_attn.qkv_proj.k_proj.weight8
8
input83<02>2%layers.7.self_attn.q_layernorm.weight8
>
input84<04><04>2)layers.7.self_attn.qkv_proj.q_proj.weight8
2
input85<04><10> 2layers.7.mlp.down_proj.weight8
0
input86<04> <0C>2layers.7.mlp.up_proj.weight8
;
input87<02>2(layers.7.post_attention_layernorm.weight8
2
input88<04> <0C>2layers.7.mlp.gate_proj.weight8
<
input89<04><10>2'layers.8.self_attn.o_proj.o_proj.weight8
>
input90<04><02>2)layers.8.self_attn.qkv_proj.v_proj.weight8
2
input91<02>2layers.8.input_layernorm.weight8
8
input92<02>2%layers.8.self_attn.k_layernorm.weight8
>
input93<04><02>2)layers.8.self_attn.qkv_proj.k_proj.weight8
8
input94<02>2%layers.8.self_attn.q_layernorm.weight8
>
input95<04><04>2)layers.8.self_attn.qkv_proj.q_proj.weight8
2
input96<04><10> 2layers.8.mlp.down_proj.weight8
0
input97<04> <0C>2layers.8.mlp.up_proj.weight8
;
input98<02>2(layers.8.post_attention_layernorm.weight8
2
input99<04> <0C>2layers.8.mlp.gate_proj.weight8
=
input100<04><10>2'layers.9.self_attn.o_proj.o_proj.weight8
?
input101<04><02>2)layers.9.self_attn.qkv_proj.v_proj.weight8
3
input102<02>2layers.9.input_layernorm.weight8
9
input103<02>2%layers.9.self_attn.k_layernorm.weight8
?
input104<04><02>2)layers.9.self_attn.qkv_proj.k_proj.weight8
9
input105<02>2%layers.9.self_attn.q_layernorm.weight8
?
input106<04><04>2)layers.9.self_attn.qkv_proj.q_proj.weight8
3
input107<04><10> 2layers.9.mlp.down_proj.weight8
1
input108<04> <0C>2layers.9.mlp.up_proj.weight8
<
input109<02>2(layers.9.post_attention_layernorm.weight8
3
input110<04> <0C>2layers.9.mlp.gate_proj.weight8
>
input111<04><10>2(layers.10.self_attn.o_proj.o_proj.weight8
@
input112<04><02>2*layers.10.self_attn.qkv_proj.v_proj.weight8
4
input113<02>2 layers.10.input_layernorm.weight8
:
input114<02>2&layers.10.self_attn.k_layernorm.weight8
@
input115<04><02>2*layers.10.self_attn.qkv_proj.k_proj.weight8
:
input116<02>2&layers.10.self_attn.q_layernorm.weight8
@
input117<04><04>2*layers.10.self_attn.qkv_proj.q_proj.weight8
4
input118<04><10> 2layers.10.mlp.down_proj.weight8
2
input119<04> <0C>2layers.10.mlp.up_proj.weight8
=
input120<02>2)layers.10.post_attention_layernorm.weight8
4
input121<04> <0C>2layers.10.mlp.gate_proj.weight8
>
input122<04><10>2(layers.11.self_attn.o_proj.o_proj.weight8
@
input123<04><02>2*layers.11.self_attn.qkv_proj.v_proj.weight8
4
input124<02>2 layers.11.input_layernorm.weight8
:
input125<02>2&layers.11.self_attn.k_layernorm.weight8
@
input126<04><02>2*layers.11.self_attn.qkv_proj.k_proj.weight8
:
input127<02>2&layers.11.self_attn.q_layernorm.weight8
@
input128<04><04>2*layers.11.self_attn.qkv_proj.q_proj.weight8
4
input129<04><10> 2layers.11.mlp.down_proj.weight8
2
input130<04> <0C>2layers.11.mlp.up_proj.weight8
=
input131<02>2)layers.11.post_attention_layernorm.weight8
4
input132<04> <0C>2layers.11.mlp.gate_proj.weight8
>
input133<04><10>2(layers.12.self_attn.o_proj.o_proj.weight8
@
input134<04><02>2*layers.12.self_attn.qkv_proj.v_proj.weight8
4
input135<02>2 layers.12.input_layernorm.weight8
:
input136<02>2&layers.12.self_attn.k_layernorm.weight8
@
input137<04><02>2*layers.12.self_attn.qkv_proj.k_proj.weight8
:
input138<02>2&layers.12.self_attn.q_layernorm.weight8
@
input139<04><04>2*layers.12.self_attn.qkv_proj.q_proj.weight8
4
input140<04><10> 2layers.12.mlp.down_proj.weight8
2
input141<04> <0C>2layers.12.mlp.up_proj.weight8
=
input142<02>2)layers.12.post_attention_layernorm.weight8
4
input143<04> <0C>2layers.12.mlp.gate_proj.weight8
>
input144<04><10>2(layers.13.self_attn.o_proj.o_proj.weight8
@
input145<04><02>2*layers.13.self_attn.qkv_proj.v_proj.weight8
4
input146<02>2 layers.13.input_layernorm.weight8
:
input147<02>2&layers.13.self_attn.k_layernorm.weight8
@
input148<04><02>2*layers.13.self_attn.qkv_proj.k_proj.weight8
:
input149<02>2&layers.13.self_attn.q_layernorm.weight8
@
input150<04><04>2*layers.13.self_attn.qkv_proj.q_proj.weight8
4
input151<04><10> 2layers.13.mlp.down_proj.weight8
2
input152<04> <0C>2layers.13.mlp.up_proj.weight8
=
input153<02>2)layers.13.post_attention_layernorm.weight8
4
input154<04> <0C>2layers.13.mlp.gate_proj.weight8
>
input155<04><10>2(layers.14.self_attn.o_proj.o_proj.weight8
@
input156<04><02>2*layers.14.self_attn.qkv_proj.v_proj.weight8
4
input157<02>2 layers.14.input_layernorm.weight8
:
input158<02>2&layers.14.self_attn.k_layernorm.weight8
@
input159<04><02>2*layers.14.self_attn.qkv_proj.k_proj.weight8
:
input160<02>2&layers.14.self_attn.q_layernorm.weight8
@
input161<04><04>2*layers.14.self_attn.qkv_proj.q_proj.weight8
4
input162<04><10> 2layers.14.mlp.down_proj.weight8
2
input163<04> <0C>2layers.14.mlp.up_proj.weight8
=
input164<02>2)layers.14.post_attention_layernorm.weight8
4
input165<04> <0C>2layers.14.mlp.gate_proj.weight8
>
input166<04><10>2(layers.15.self_attn.o_proj.o_proj.weight8
@
input167<04><02>2*layers.15.self_attn.qkv_proj.v_proj.weight8
4
input168<02>2 layers.15.input_layernorm.weight8
:
input169<02>2&layers.15.self_attn.k_layernorm.weight8
@
input170<04><02>2*layers.15.self_attn.qkv_proj.k_proj.weight8
:
input171<02>2&layers.15.self_attn.q_layernorm.weight8
@
input172<04><04>2*layers.15.self_attn.qkv_proj.q_proj.weight8
4
input173<04><10> 2layers.15.mlp.down_proj.weight8
2
input174<04> <0C>2layers.15.mlp.up_proj.weight8
=
input175<02>2)layers.15.post_attention_layernorm.weight8
4
input176<04> <0C>2layers.15.mlp.gate_proj.weight8
>
input177<04><10>2(layers.16.self_attn.o_proj.o_proj.weight8
@
input178<04><02>2*layers.16.self_attn.qkv_proj.v_proj.weight8
4
input179<02>2 layers.16.input_layernorm.weight8
:
input180<02>2&layers.16.self_attn.k_layernorm.weight8
@
input181<04><02>2*layers.16.self_attn.qkv_proj.k_proj.weight8
:
input182<02>2&layers.16.self_attn.q_layernorm.weight8
@
input183<04><04>2*layers.16.self_attn.qkv_proj.q_proj.weight8
4
input184<04><10> 2layers.16.mlp.down_proj.weight8
2
input185<04> <0C>2layers.16.mlp.up_proj.weight8
=
input186<02>2)layers.16.post_attention_layernorm.weight8
4
input187<04> <0C>2layers.16.mlp.gate_proj.weight8
>
input188<04><10>2(layers.17.self_attn.o_proj.o_proj.weight8
@
input189<04><02>2*layers.17.self_attn.qkv_proj.v_proj.weight8
4
input190<02>2 layers.17.input_layernorm.weight8
:
input191<02>2&layers.17.self_attn.k_layernorm.weight8
@
input192<04><02>2*layers.17.self_attn.qkv_proj.k_proj.weight8
:
input193<02>2&layers.17.self_attn.q_layernorm.weight8
@
input194<04><04>2*layers.17.self_attn.qkv_proj.q_proj.weight8
4
input195<04><10> 2layers.17.mlp.down_proj.weight8
2
input196<04> <0C>2layers.17.mlp.up_proj.weight8
=
input197<02>2)layers.17.post_attention_layernorm.weight8
4
input198<04> <0C>2layers.17.mlp.gate_proj.weight8
>
input199<04><10>2(layers.18.self_attn.o_proj.o_proj.weight8
@
input200<04><02>2*layers.18.self_attn.qkv_proj.v_proj.weight8
4
input201<02>2 layers.18.input_layernorm.weight8
:
input202<02>2&layers.18.self_attn.k_layernorm.weight8
@
input203<04><02>2*layers.18.self_attn.qkv_proj.k_proj.weight8
:
input204<02>2&layers.18.self_attn.q_layernorm.weight8
@
input205<04><04>2*layers.18.self_attn.qkv_proj.q_proj.weight8
4
input206<04><10> 2layers.18.mlp.down_proj.weight8
2
input207<04> <0C>2layers.18.mlp.up_proj.weight8
=
input208<02>2)layers.18.post_attention_layernorm.weight8
4
input209<04> <0C>2layers.18.mlp.gate_proj.weight8
>
input210<04><10>2(layers.19.self_attn.o_proj.o_proj.weight8
@
input211<04><02>2*layers.19.self_attn.qkv_proj.v_proj.weight8
4
input212<02>2 layers.19.input_layernorm.weight8
:
input213<02>2&layers.19.self_attn.k_layernorm.weight8
@
input214<04><02>2*layers.19.self_attn.qkv_proj.k_proj.weight8
:
input215<02>2&layers.19.self_attn.q_layernorm.weight8
@
input216<04><04>2*layers.19.self_attn.qkv_proj.q_proj.weight8
4
input217<04><10> 2layers.19.mlp.down_proj.weight8
2
input218<04> <0C>2layers.19.mlp.up_proj.weight8
=
input219<02>2)layers.19.post_attention_layernorm.weight8
4
input220<04> <0C>2layers.19.mlp.gate_proj.weight8
>
input221<04><10>2(layers.20.self_attn.o_proj.o_proj.weight8
@
input222<04><02>2*layers.20.self_attn.qkv_proj.v_proj.weight8
4
input223<02>2 layers.20.input_layernorm.weight8
:
input224<02>2&layers.20.self_attn.k_layernorm.weight8
@
input225<04><02>2*layers.20.self_attn.qkv_proj.k_proj.weight8
:
input226<02>2&layers.20.self_attn.q_layernorm.weight8
@
input227<04><04>2*layers.20.self_attn.qkv_proj.q_proj.weight8
4
input228<04><10> 2layers.20.mlp.down_proj.weight8
2
input229<04> <0C>2layers.20.mlp.up_proj.weight8
=
input230<02>2)layers.20.post_attention_layernorm.weight8
4
input231<04> <0C>2layers.20.mlp.gate_proj.weight8
>
input232<04><10>2(layers.21.self_attn.o_proj.o_proj.weight8
@
input233<04><02>2*layers.21.self_attn.qkv_proj.v_proj.weight8
4
input234<02>2 layers.21.input_layernorm.weight8
:
input235<02>2&layers.21.self_attn.k_layernorm.weight8
@
input236<04><02>2*layers.21.self_attn.qkv_proj.k_proj.weight8
:
input237<02>2&layers.21.self_attn.q_layernorm.weight8
@
input238<04><04>2*layers.21.self_attn.qkv_proj.q_proj.weight8
4
input239<04><10> 2layers.21.mlp.down_proj.weight8
2
input240<04> <0C>2layers.21.mlp.up_proj.weight8
=
input241<02>2)layers.21.post_attention_layernorm.weight8
4
input242<04> <0C>2layers.21.mlp.gate_proj.weight8
>
input243<04><10>2(layers.22.self_attn.o_proj.o_proj.weight8
@
input244<04><02>2*layers.22.self_attn.qkv_proj.v_proj.weight8
4
input245<02>2 layers.22.input_layernorm.weight8
:
input246<02>2&layers.22.self_attn.k_layernorm.weight8
@
input247<04><02>2*layers.22.self_attn.qkv_proj.k_proj.weight8
:
input248<02>2&layers.22.self_attn.q_layernorm.weight8
@
input249<04><04>2*layers.22.self_attn.qkv_proj.q_proj.weight8
4
input250<04><10> 2layers.22.mlp.down_proj.weight8
2
input251<04> <0C>2layers.22.mlp.up_proj.weight8
=
input252<02>2)layers.22.post_attention_layernorm.weight8
4
input253<04> <0C>2layers.22.mlp.gate_proj.weight8
>
input254<04><10>2(layers.23.self_attn.o_proj.o_proj.weight8
@
input255<04><02>2*layers.23.self_attn.qkv_proj.v_proj.weight8
4
input256<02>2 layers.23.input_layernorm.weight8
:
input257<02>2&layers.23.self_attn.k_layernorm.weight8
@
input258<04><02>2*layers.23.self_attn.qkv_proj.k_proj.weight8
:
input259<02>2&layers.23.self_attn.q_layernorm.weight8
@
input260<04><04>2*layers.23.self_attn.qkv_proj.q_proj.weight8
4
input261<04><10> 2layers.23.mlp.down_proj.weight8
2
input262<04> <0C>2layers.23.mlp.up_proj.weight8
=
input263<02>2)layers.23.post_attention_layernorm.weight8
4
input264<04> <0C>2layers.23.mlp.gate_proj.weight8
>
input265<04><10>2(layers.24.self_attn.o_proj.o_proj.weight8
@
input266<04><02>2*layers.24.self_attn.qkv_proj.v_proj.weight8
4
input267<02>2 layers.24.input_layernorm.weight8
:
input268<02>2&layers.24.self_attn.k_layernorm.weight8
@
input269<04><02>2*layers.24.self_attn.qkv_proj.k_proj.weight8
:
input270<02>2&layers.24.self_attn.q_layernorm.weight8
@
input271<04><04>2*layers.24.self_attn.qkv_proj.q_proj.weight8
4
input272<04><10> 2layers.24.mlp.down_proj.weight8
2
input273<04> <0C>2layers.24.mlp.up_proj.weight8
=
input274<02>2)layers.24.post_attention_layernorm.weight8
4
input275<04> <0C>2layers.24.mlp.gate_proj.weight8
>
input276<04><10>2(layers.25.self_attn.o_proj.o_proj.weight8
@
input277<04><02>2*layers.25.self_attn.qkv_proj.v_proj.weight8
4
input278<02>2 layers.25.input_layernorm.weight8
:
input279<02>2&layers.25.self_attn.k_layernorm.weight8
@
input280<04><02>2*layers.25.self_attn.qkv_proj.k_proj.weight8
:
input281<02>2&layers.25.self_attn.q_layernorm.weight8
@
input282<04><04>2*layers.25.self_attn.qkv_proj.q_proj.weight8
4
input283<04><10> 2layers.25.mlp.down_proj.weight8
2
input284<04> <0C>2layers.25.mlp.up_proj.weight8
=
input285<02>2)layers.25.post_attention_layernorm.weight8
4
input286<04> <0C>2layers.25.mlp.gate_proj.weight8
>
input287<04><10>2(layers.26.self_attn.o_proj.o_proj.weight8
@
input288<04><02>2*layers.26.self_attn.qkv_proj.v_proj.weight8
4
input289<02>2 layers.26.input_layernorm.weight8
:
input290<02>2&layers.26.self_attn.k_layernorm.weight8
@
input291<04><02>2*layers.26.self_attn.qkv_proj.k_proj.weight8
:
input292<02>2&layers.26.self_attn.q_layernorm.weight8
@
input293<04><04>2*layers.26.self_attn.qkv_proj.q_proj.weight8
4
input294<04><10> 2layers.26.mlp.down_proj.weight8
2
input295<04> <0C>2layers.26.mlp.up_proj.weight8
=
input296<02>2)layers.26.post_attention_layernorm.weight8
4
input297<04> <0C>2layers.26.mlp.gate_proj.weight8
>
input298<04><10>2(layers.27.self_attn.o_proj.o_proj.weight8
@
input299<04><02>2*layers.27.self_attn.qkv_proj.v_proj.weight8
4
input300<02>2 layers.27.input_layernorm.weight8
:
input301<02>2&layers.27.self_attn.k_layernorm.weight8
@
input302<04><02>2*layers.27.self_attn.qkv_proj.k_proj.weight8
:
input303<02>2&layers.27.self_attn.q_layernorm.weight8
@
input304<04><04>2*layers.27.self_attn.qkv_proj.q_proj.weight8
4
input305<04><10> 2layers.27.mlp.down_proj.weight8
2
input306<04> <0C>2layers.27.mlp.up_proj.weight8
=
input307<02>2)layers.27.post_attention_layernorm.weight8
4
input308<04> <0C>2layers.27.mlp.gate_proj.weight8
%
input309<05><><02>2lm_head.weight8

input310<02>2 norm.weight8'
output0<05><> <09>2embed_tokens.weight:
output1<04><10>2'layers.0.self_attn.o_proj.o_proj.weight<
output2<04><02>2)layers.0.self_attn.qkv_proj.v_proj.weight0
output3<02>2layers.0.input_layernorm.weight6
output4<02>2%layers.0.self_attn.k_layernorm.weight<
output5<04><02>2)layers.0.self_attn.qkv_proj.k_proj.weight6
output6<02>2%layers.0.self_attn.q_layernorm.weight<
output7<04><04>2)layers.0.self_attn.qkv_proj.q_proj.weight0
output8<04><10> 2layers.0.mlp.down_proj.weight.
output9<04> <0C>2layers.0.mlp.up_proj.weight:
output10<02>2(layers.0.post_attention_layernorm.weight1
output11<04> <0C>2layers.0.mlp.gate_proj.weight;
output12<04><10>2'layers.1.self_attn.o_proj.o_proj.weight=
output13<04><02>2)layers.1.self_attn.qkv_proj.v_proj.weight1
output14<02>2layers.1.input_layernorm.weight7
output15<02>2%layers.1.self_attn.k_layernorm.weight=
output16<04><02>2)layers.1.self_attn.qkv_proj.k_proj.weight7
output17<02>2%layers.1.self_attn.q_layernorm.weight=
output18<04><04>2)layers.1.self_attn.qkv_proj.q_proj.weight1
output19<04><10> 2layers.1.mlp.down_proj.weight/
output20<04> <0C>2layers.1.mlp.up_proj.weight:
output21<02>2(layers.1.post_attention_layernorm.weight1
output22<04> <0C>2layers.1.mlp.gate_proj.weight;
output23<04><10>2'layers.2.self_attn.o_proj.o_proj.weight=
output24<04><02>2)layers.2.self_attn.qkv_proj.v_proj.weight1
output25<02>2layers.2.input_layernorm.weight7
output26<02>2%layers.2.self_attn.k_layernorm.weight=
output27<04><02>2)layers.2.self_attn.qkv_proj.k_proj.weight7
output28<02>2%layers.2.self_attn.q_layernorm.weight=
output29<04><04>2)layers.2.self_attn.qkv_proj.q_proj.weight1
output30<04><10> 2layers.2.mlp.down_proj.weight/
output31<04> <0C>2layers.2.mlp.up_proj.weight:
output32<02>2(layers.2.post_attention_layernorm.weight1
output33<04> <0C>2layers.2.mlp.gate_proj.weight;
output34<04><10>2'layers.3.self_attn.o_proj.o_proj.weight=
output35<04><02>2)layers.3.self_attn.qkv_proj.v_proj.weight1
output36<02>2layers.3.input_layernorm.weight7
output37<02>2%layers.3.self_attn.k_layernorm.weight=
output38<04><02>2)layers.3.self_attn.qkv_proj.k_proj.weight7
output39<02>2%layers.3.self_attn.q_layernorm.weight=
output40<04><04>2)layers.3.self_attn.qkv_proj.q_proj.weight1
output41<04><10> 2layers.3.mlp.down_proj.weight/
output42<04> <0C>2layers.3.mlp.up_proj.weight:
output43<02>2(layers.3.post_attention_layernorm.weight1
output44<04> <0C>2layers.3.mlp.gate_proj.weight;
output45<04><10>2'layers.4.self_attn.o_proj.o_proj.weight=
output46<04><02>2)layers.4.self_attn.qkv_proj.v_proj.weight1
output47<02>2layers.4.input_layernorm.weight7
output48<02>2%layers.4.self_attn.k_layernorm.weight=
output49<04><02>2)layers.4.self_attn.qkv_proj.k_proj.weight7
output50<02>2%layers.4.self_attn.q_layernorm.weight=
output51<04><04>2)layers.4.self_attn.qkv_proj.q_proj.weight1
output52<04><10> 2layers.4.mlp.down_proj.weight/
output53<04> <0C>2layers.4.mlp.up_proj.weight:
output54<02>2(layers.4.post_attention_layernorm.weight1
output55<04> <0C>2layers.4.mlp.gate_proj.weight;
output56<04><10>2'layers.5.self_attn.o_proj.o_proj.weight=
output57<04><02>2)layers.5.self_attn.qkv_proj.v_proj.weight1
output58<02>2layers.5.input_layernorm.weight7
output59<02>2%layers.5.self_attn.k_layernorm.weight=
output60<04><02>2)layers.5.self_attn.qkv_proj.k_proj.weight7
output61<02>2%layers.5.self_attn.q_layernorm.weight=
output62<04><04>2)layers.5.self_attn.qkv_proj.q_proj.weight1
output63<04><10> 2layers.5.mlp.down_proj.weight/
output64<04> <0C>2layers.5.mlp.up_proj.weight:
output65<02>2(layers.5.post_attention_layernorm.weight1
output66<04> <0C>2layers.5.mlp.gate_proj.weight;
output67<04><10>2'layers.6.self_attn.o_proj.o_proj.weight=
output68<04><02>2)layers.6.self_attn.qkv_proj.v_proj.weight1
output69<02>2layers.6.input_layernorm.weight7
output70<02>2%layers.6.self_attn.k_layernorm.weight=
output71<04><02>2)layers.6.self_attn.qkv_proj.k_proj.weight7
output72<02>2%layers.6.self_attn.q_layernorm.weight=
output73<04><04>2)layers.6.self_attn.qkv_proj.q_proj.weight1
output74<04><10> 2layers.6.mlp.down_proj.weight/
output75<04> <0C>2layers.6.mlp.up_proj.weight:
output76<02>2(layers.6.post_attention_layernorm.weight1
output77<04> <0C>2layers.6.mlp.gate_proj.weight;
output78<04><10>2'layers.7.self_attn.o_proj.o_proj.weight=
output79<04><02>2)layers.7.self_attn.qkv_proj.v_proj.weight1
output80<02>2layers.7.input_layernorm.weight7
output81<02>2%layers.7.self_attn.k_layernorm.weight=
output82<04><02>2)layers.7.self_attn.qkv_proj.k_proj.weight7
output83<02>2%layers.7.self_attn.q_layernorm.weight=
output84<04><04>2)layers.7.self_attn.qkv_proj.q_proj.weight1
output85<04><10> 2layers.7.mlp.down_proj.weight/
output86<04> <0C>2layers.7.mlp.up_proj.weight:
output87<02>2(layers.7.post_attention_layernorm.weight1
output88<04> <0C>2layers.7.mlp.gate_proj.weight;
output89<04><10>2'layers.8.self_attn.o_proj.o_proj.weight=
output90<04><02>2)layers.8.self_attn.qkv_proj.v_proj.weight1
output91<02>2layers.8.input_layernorm.weight7
output92<02>2%layers.8.self_attn.k_layernorm.weight=
output93<04><02>2)layers.8.self_attn.qkv_proj.k_proj.weight7
output94<02>2%layers.8.self_attn.q_layernorm.weight=
output95<04><04>2)layers.8.self_attn.qkv_proj.q_proj.weight1
output96<04><10> 2layers.8.mlp.down_proj.weight/
output97<04> <0C>2layers.8.mlp.up_proj.weight:
output98<02>2(layers.8.post_attention_layernorm.weight1
output99<04> <0C>2layers.8.mlp.gate_proj.weight<
output100<04><10>2'layers.9.self_attn.o_proj.o_proj.weight>
output101<04><02>2)layers.9.self_attn.qkv_proj.v_proj.weight2
output102<02>2layers.9.input_layernorm.weight8
output103<02>2%layers.9.self_attn.k_layernorm.weight>
output104<04><02>2)layers.9.self_attn.qkv_proj.k_proj.weight8
output105<02>2%layers.9.self_attn.q_layernorm.weight>
output106<04><04>2)layers.9.self_attn.qkv_proj.q_proj.weight2
output107<04><10> 2layers.9.mlp.down_proj.weight0
output108<04> <0C>2layers.9.mlp.up_proj.weight;
output109<02>2(layers.9.post_attention_layernorm.weight2
output110<04> <0C>2layers.9.mlp.gate_proj.weight=
output111<04><10>2(layers.10.self_attn.o_proj.o_proj.weight?
output112<04><02>2*layers.10.self_attn.qkv_proj.v_proj.weight3
output113<02>2 layers.10.input_layernorm.weight9
output114<02>2&layers.10.self_attn.k_layernorm.weight?
output115<04><02>2*layers.10.self_attn.qkv_proj.k_proj.weight9
output116<02>2&layers.10.self_attn.q_layernorm.weight?
output117<04><04>2*layers.10.self_attn.qkv_proj.q_proj.weight3
output118<04><10> 2layers.10.mlp.down_proj.weight1
output119<04> <0C>2layers.10.mlp.up_proj.weight<
output120<02>2)layers.10.post_attention_layernorm.weight3
output121<04> <0C>2layers.10.mlp.gate_proj.weight=
output122<04><10>2(layers.11.self_attn.o_proj.o_proj.weight?
output123<04><02>2*layers.11.self_attn.qkv_proj.v_proj.weight3
output124<02>2 layers.11.input_layernorm.weight9
output125<02>2&layers.11.self_attn.k_layernorm.weight?
output126<04><02>2*layers.11.self_attn.qkv_proj.k_proj.weight9
output127<02>2&layers.11.self_attn.q_layernorm.weight?
output128<04><04>2*layers.11.self_attn.qkv_proj.q_proj.weight3
output129<04><10> 2layers.11.mlp.down_proj.weight1
output130<04> <0C>2layers.11.mlp.up_proj.weight<
output131<02>2)layers.11.post_attention_layernorm.weight3
output132<04> <0C>2layers.11.mlp.gate_proj.weight=
output133<04><10>2(layers.12.self_attn.o_proj.o_proj.weight?
output134<04><02>2*layers.12.self_attn.qkv_proj.v_proj.weight3
output135<02>2 layers.12.input_layernorm.weight9
output136<02>2&layers.12.self_attn.k_layernorm.weight?
output137<04><02>2*layers.12.self_attn.qkv_proj.k_proj.weight9
output138<02>2&layers.12.self_attn.q_layernorm.weight?
output139<04><04>2*layers.12.self_attn.qkv_proj.q_proj.weight3
output140<04><10> 2layers.12.mlp.down_proj.weight1
output141<04> <0C>2layers.12.mlp.up_proj.weight<
output142<02>2)layers.12.post_attention_layernorm.weight3
output143<04> <0C>2layers.12.mlp.gate_proj.weight=
output144<04><10>2(layers.13.self_attn.o_proj.o_proj.weight?
output145<04><02>2*layers.13.self_attn.qkv_proj.v_proj.weight3
output146<02>2 layers.13.input_layernorm.weight9
output147<02>2&layers.13.self_attn.k_layernorm.weight?
output148<04><02>2*layers.13.self_attn.qkv_proj.k_proj.weight9
output149<02>2&layers.13.self_attn.q_layernorm.weight?
output150<04><04>2*layers.13.self_attn.qkv_proj.q_proj.weight3
output151<04><10> 2layers.13.mlp.down_proj.weight1
output152<04> <0C>2layers.13.mlp.up_proj.weight<
output153<02>2)layers.13.post_attention_layernorm.weight3
output154<04> <0C>2layers.13.mlp.gate_proj.weight=
output155<04><10>2(layers.14.self_attn.o_proj.o_proj.weight?
output156<04><02>2*layers.14.self_attn.qkv_proj.v_proj.weight3
output157<02>2 layers.14.input_layernorm.weight9
output158<02>2&layers.14.self_attn.k_layernorm.weight?
output159<04><02>2*layers.14.self_attn.qkv_proj.k_proj.weight9
output160<02>2&layers.14.self_attn.q_layernorm.weight?
output161<04><04>2*layers.14.self_attn.qkv_proj.q_proj.weight3
output162<04><10> 2layers.14.mlp.down_proj.weight1
output163<04> <0C>2layers.14.mlp.up_proj.weight<
output164<02>2)layers.14.post_attention_layernorm.weight3
output165<04> <0C>2layers.14.mlp.gate_proj.weight=
output166<04><10>2(layers.15.self_attn.o_proj.o_proj.weight?
output167<04><02>2*layers.15.self_attn.qkv_proj.v_proj.weight3
output168<02>2 layers.15.input_layernorm.weight9
output169<02>2&layers.15.self_attn.k_layernorm.weight?
output170<04><02>2*layers.15.self_attn.qkv_proj.k_proj.weight9
output171<02>2&layers.15.self_attn.q_layernorm.weight?
output172<04><04>2*layers.15.self_attn.qkv_proj.q_proj.weight3
output173<04><10> 2layers.15.mlp.down_proj.weight1
output174<04> <0C>2layers.15.mlp.up_proj.weight<
output175<02>2)layers.15.post_attention_layernorm.weight3
output176<04> <0C>2layers.15.mlp.gate_proj.weight=
output177<04><10>2(layers.16.self_attn.o_proj.o_proj.weight?
output178<04><02>2*layers.16.self_attn.qkv_proj.v_proj.weight3
output179<02>2 layers.16.input_layernorm.weight9
output180<02>2&layers.16.self_attn.k_layernorm.weight?
output181<04><02>2*layers.16.self_attn.qkv_proj.k_proj.weight9
output182<02>2&layers.16.self_attn.q_layernorm.weight?
output183<04><04>2*layers.16.self_attn.qkv_proj.q_proj.weight3
output184<04><10> 2layers.16.mlp.down_proj.weight1
output185<04> <0C>2layers.16.mlp.up_proj.weight<
output186<02>2)layers.16.post_attention_layernorm.weight3
output187<04> <0C>2layers.16.mlp.gate_proj.weight=
output188<04><10>2(layers.17.self_attn.o_proj.o_proj.weight?
output189<04><02>2*layers.17.self_attn.qkv_proj.v_proj.weight3
output190<02>2 layers.17.input_layernorm.weight9
output191<02>2&layers.17.self_attn.k_layernorm.weight?
output192<04><02>2*layers.17.self_attn.qkv_proj.k_proj.weight9
output193<02>2&layers.17.self_attn.q_layernorm.weight?
output194<04><04>2*layers.17.self_attn.qkv_proj.q_proj.weight3
output195<04><10> 2layers.17.mlp.down_proj.weight1
output196<04> <0C>2layers.17.mlp.up_proj.weight<
output197<02>2)layers.17.post_attention_layernorm.weight3
output198<04> <0C>2layers.17.mlp.gate_proj.weight=
output199<04><10>2(layers.18.self_attn.o_proj.o_proj.weight?
output200<04><02>2*layers.18.self_attn.qkv_proj.v_proj.weight3
output201<02>2 layers.18.input_layernorm.weight9
output202<02>2&layers.18.self_attn.k_layernorm.weight?
output203<04><02>2*layers.18.self_attn.qkv_proj.k_proj.weight9
output204<02>2&layers.18.self_attn.q_layernorm.weight?
output205<04><04>2*layers.18.self_attn.qkv_proj.q_proj.weight3
output206<04><10> 2layers.18.mlp.down_proj.weight1
output207<04> <0C>2layers.18.mlp.up_proj.weight<
output208<02>2)layers.18.post_attention_layernorm.weight3
output209<04> <0C>2layers.18.mlp.gate_proj.weight=
output210<04><10>2(layers.19.self_attn.o_proj.o_proj.weight?
output211<04><02>2*layers.19.self_attn.qkv_proj.v_proj.weight3
output212<02>2 layers.19.input_layernorm.weight9
output213<02>2&layers.19.self_attn.k_layernorm.weight?
output214<04><02>2*layers.19.self_attn.qkv_proj.k_proj.weight9
output215<02>2&layers.19.self_attn.q_layernorm.weight?
output216<04><04>2*layers.19.self_attn.qkv_proj.q_proj.weight3
output217<04><10> 2layers.19.mlp.down_proj.weight1
output218<04> <0C>2layers.19.mlp.up_proj.weight<
output219<02>2)layers.19.post_attention_layernorm.weight3
output220<04> <0C>2layers.19.mlp.gate_proj.weight=
output221<04><10>2(layers.20.self_attn.o_proj.o_proj.weight?
output222<04><02>2*layers.20.self_attn.qkv_proj.v_proj.weight3
output223<02>2 layers.20.input_layernorm.weight9
output224<02>2&layers.20.self_attn.k_layernorm.weight?
output225<04><02>2*layers.20.self_attn.qkv_proj.k_proj.weight9
output226<02>2&layers.20.self_attn.q_layernorm.weight?
output227<04><04>2*layers.20.self_attn.qkv_proj.q_proj.weight3
output228<04><10> 2layers.20.mlp.down_proj.weight1
output229<04> <0C>2layers.20.mlp.up_proj.weight<
output230<02>2)layers.20.post_attention_layernorm.weight3
output231<04> <0C>2layers.20.mlp.gate_proj.weight=
output232<04><10>2(layers.21.self_attn.o_proj.o_proj.weight?
output233<04><02>2*layers.21.self_attn.qkv_proj.v_proj.weight3
output234<02>2 layers.21.input_layernorm.weight9
output235<02>2&layers.21.self_attn.k_layernorm.weight?
output236<04><02>2*layers.21.self_attn.qkv_proj.k_proj.weight9
output237<02>2&layers.21.self_attn.q_layernorm.weight?
output238<04><04>2*layers.21.self_attn.qkv_proj.q_proj.weight3
output239<04><10> 2layers.21.mlp.down_proj.weight1
output240<04> <0C>2layers.21.mlp.up_proj.weight<
output241<02>2)layers.21.post_attention_layernorm.weight3
output242<04> <0C>2layers.21.mlp.gate_proj.weight=
output243<04><10>2(layers.22.self_attn.o_proj.o_proj.weight?
output244<04><02>2*layers.22.self_attn.qkv_proj.v_proj.weight3
output245<02>2 layers.22.input_layernorm.weight9
output246<02>2&layers.22.self_attn.k_layernorm.weight?
output247<04><02>2*layers.22.self_attn.qkv_proj.k_proj.weight9
output248<02>2&layers.22.self_attn.q_layernorm.weight?
output249<04><04>2*layers.22.self_attn.qkv_proj.q_proj.weight3
output250<04><10> 2layers.22.mlp.down_proj.weight1
output251<04> <0C>2layers.22.mlp.up_proj.weight<
output252<02>2)layers.22.post_attention_layernorm.weight3
output253<04> <0C>2layers.22.mlp.gate_proj.weight=
output254<04><10>2(layers.23.self_attn.o_proj.o_proj.weight?
output255<04><02>2*layers.23.self_attn.qkv_proj.v_proj.weight3
output256<02>2 layers.23.input_layernorm.weight9
output257<02>2&layers.23.self_attn.k_layernorm.weight?
output258<04><02>2*layers.23.self_attn.qkv_proj.k_proj.weight9
output259<02>2&layers.23.self_attn.q_layernorm.weight?
output260<04><04>2*layers.23.self_attn.qkv_proj.q_proj.weight3
output261<04><10> 2layers.23.mlp.down_proj.weight1
output262<04> <0C>2layers.23.mlp.up_proj.weight<
output263<02>2)layers.23.post_attention_layernorm.weight3
output264<04> <0C>2layers.23.mlp.gate_proj.weight=
output265<04><10>2(layers.24.self_attn.o_proj.o_proj.weight?
output266<04><02>2*layers.24.self_attn.qkv_proj.v_proj.weight3
output267<02>2 layers.24.input_layernorm.weight9
output268<02>2&layers.24.self_attn.k_layernorm.weight?
output269<04><02>2*layers.24.self_attn.qkv_proj.k_proj.weight9
output270<02>2&layers.24.self_attn.q_layernorm.weight?
output271<04><04>2*layers.24.self_attn.qkv_proj.q_proj.weight3
output272<04><10> 2layers.24.mlp.down_proj.weight1
output273<04> <0C>2layers.24.mlp.up_proj.weight<
output274<02>2)layers.24.post_attention_layernorm.weight3
output275<04> <0C>2layers.24.mlp.gate_proj.weight=
output276<04><10>2(layers.25.self_attn.o_proj.o_proj.weight?
output277<04><02>2*layers.25.self_attn.qkv_proj.v_proj.weight3
output278<02>2 layers.25.input_layernorm.weight9
output279<02>2&layers.25.self_attn.k_layernorm.weight?
output280<04><02>2*layers.25.self_attn.qkv_proj.k_proj.weight9
output281<02>2&layers.25.self_attn.q_layernorm.weight?
output282<04><04>2*layers.25.self_attn.qkv_proj.q_proj.weight3
output283<04><10> 2layers.25.mlp.down_proj.weight1
output284<04> <0C>2layers.25.mlp.up_proj.weight<
output285<02>2)layers.25.post_attention_layernorm.weight3
output286<04> <0C>2layers.25.mlp.gate_proj.weight=
output287<04><10>2(layers.26.self_attn.o_proj.o_proj.weight?
output288<04><02>2*layers.26.self_attn.qkv_proj.v_proj.weight3
output289<02>2 layers.26.input_layernorm.weight9
output290<02>2&layers.26.self_attn.k_layernorm.weight?
output291<04><02>2*layers.26.self_attn.qkv_proj.k_proj.weight9
output292<02>2&layers.26.self_attn.q_layernorm.weight?
output293<04><04>2*layers.26.self_attn.qkv_proj.q_proj.weight3
output294<04><10> 2layers.26.mlp.down_proj.weight1
output295<04> <0C>2layers.26.mlp.up_proj.weight<
output296<02>2)layers.26.post_attention_layernorm.weight3
output297<04> <0C>2layers.26.mlp.gate_proj.weight=
output298<04><10>2(layers.27.self_attn.o_proj.o_proj.weight?
output299<04><02>2*layers.27.self_attn.qkv_proj.v_proj.weight3
output300<02>2 layers.27.input_layernorm.weight9
output301<02>2&layers.27.self_attn.k_layernorm.weight?
output302<04><02>2*layers.27.self_attn.qkv_proj.k_proj.weight9
output303<02>2&layers.27.self_attn.q_layernorm.weight?
output304<04><04>2*layers.27.self_attn.qkv_proj.q_proj.weight3
output305<04><10> 2layers.27.mlp.down_proj.weight1
output306<04> <0C>2layers.27.mlp.up_proj.weight<
output307<02>2)layers.27.post_attention_layernorm.weight3
output308<04> <0C>2layers.27.mlp.gate_proj.weight$
output309<05><><02>2lm_head.weight
output310<02>2 norm.weight