Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- 2b_l10/outputs_add_free/all/lang/0/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/101/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/102/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/11/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/113/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/117/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/120/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/121/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/122/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/123/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/125/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/141/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/142/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/146/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/151/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/169/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/170/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/176/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/184/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/193/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/2/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/20/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/205/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/211/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/213/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/217/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/218/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/221/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/228/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/23/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/239/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/25/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/254/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/256/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/257/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/261/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/262/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/267/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/272/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/275/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/278/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/280/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/282/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/285/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/295/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/296/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/297/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/299/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/303/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/305/config.json +12 -0
2b_l10/outputs_add_free/all/lang/0/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/101/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/102/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/11/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/113/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/117/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/120/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/121/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/122/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/123/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/125/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/141/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/142/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/146/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/151/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/169/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/170/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/176/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/184/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/193/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/2/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/20/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/205/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/211/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/213/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/217/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/218/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/221/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/228/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/23/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/239/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/25/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/254/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/256/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/257/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/261/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/262/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/267/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/272/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/275/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/278/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/280/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/282/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/285/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/295/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/296/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/297/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/299/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/303/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/305/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|