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/106/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/109/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/110/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/127/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/130/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/138/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/139/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/148/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/155/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/156/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/16/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/162/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/171/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/172/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/177/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/183/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/186/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/19/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/194/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/201/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/202/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/206/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/212/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/214/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/216/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/225/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/227/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/238/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/243/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/244/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/249/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/26/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/27/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/270/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/277/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/279/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/281/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/283/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/284/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/290/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/312/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/324/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/325/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/327/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/337/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/341/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/342/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/347/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/358/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/384/config.json +12 -0
2b_l10/outputs_add_free/all/lang/106/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/109/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/110/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/127/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/130/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/138/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/139/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/148/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/155/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/156/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/16/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/162/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/171/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/172/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/177/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/183/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/186/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/19/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/194/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/201/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/202/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/206/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/212/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/214/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/216/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/225/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/227/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/238/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/243/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/244/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/249/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/26/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/27/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/270/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/277/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/279/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/281/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/283/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/284/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/290/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/312/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/324/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/325/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/327/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/337/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/341/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/342/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/347/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/358/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/384/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 |
+
]
|