Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-1000/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-1000/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +161 -0
- checkpoint-1200/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-1200/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-1200/optimizer.pt +3 -0
- checkpoint-1200/rng_state.pth +3 -0
- checkpoint-1200/scheduler.pt +3 -0
- checkpoint-1200/trainer_state.json +189 -0
- checkpoint-1400/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-1400/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-1400/optimizer.pt +3 -0
- checkpoint-1400/rng_state.pth +3 -0
- checkpoint-1400/scheduler.pt +3 -0
- checkpoint-1400/trainer_state.json +217 -0
- checkpoint-1600/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-1600/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-1600/optimizer.pt +3 -0
- checkpoint-1600/rng_state.pth +3 -0
- checkpoint-1600/scheduler.pt +3 -0
- checkpoint-1600/trainer_state.json +245 -0
- checkpoint-1800/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-1800/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-1800/optimizer.pt +3 -0
- checkpoint-1800/rng_state.pth +3 -0
- checkpoint-1800/scheduler.pt +3 -0
- checkpoint-1800/trainer_state.json +273 -0
- checkpoint-200/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-200/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-200/optimizer.pt +3 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/trainer_state.json +49 -0
- checkpoint-400/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-400/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-400/optimizer.pt +3 -0
- checkpoint-400/rng_state.pth +3 -0
- checkpoint-400/scheduler.pt +3 -0
- checkpoint-400/trainer_state.json +77 -0
- checkpoint-600/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-600/intervenable_model/pytorch_model.bin +3 -0
- checkpoint-600/optimizer.pt +3 -0
- checkpoint-600/rng_state.pth +3 -0
- checkpoint-600/scheduler.pt +3 -0
- checkpoint-600/trainer_state.json +105 -0
- checkpoint-800/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin +3 -0
- checkpoint-800/intervenable_model/pytorch_model.bin +3 -0
checkpoint-1000/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:748832210afa5ebf685ff723b101678a23d7cdc70dc62bccba1a68fa8430493a
|
| 3 |
+
size 100495
|
checkpoint-1000/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55b14428571fc945e6aff92b76c58b76839508fc8a5a725e81f9444eb1e2e2d9
|
| 3 |
+
size 200278
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12de1d084c8dff51a1b37fd4b9f81b3421264efab7df040134d72fc44d217e09
|
| 3 |
+
size 1064
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5252100840336135,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.34138655462184875,
|
| 97 |
+
"grad_norm": 2.95237398147583,
|
| 98 |
+
"learning_rate": 0.002634453781512605,
|
| 99 |
+
"loss": 0.9322,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36764705882352944,
|
| 104 |
+
"grad_norm": 3.39207124710083,
|
| 105 |
+
"learning_rate": 0.0025294117647058825,
|
| 106 |
+
"loss": 0.939,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3939075630252101,
|
| 111 |
+
"grad_norm": 3.5582635402679443,
|
| 112 |
+
"learning_rate": 0.0024243697478991597,
|
| 113 |
+
"loss": 1.0108,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42016806722689076,
|
| 118 |
+
"grad_norm": 3.2852275371551514,
|
| 119 |
+
"learning_rate": 0.0023193277310924374,
|
| 120 |
+
"loss": 0.884,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.44642857142857145,
|
| 125 |
+
"grad_norm": 3.8662281036376953,
|
| 126 |
+
"learning_rate": 0.0022142857142857146,
|
| 127 |
+
"loss": 0.8889,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4726890756302521,
|
| 132 |
+
"grad_norm": 3.1788485050201416,
|
| 133 |
+
"learning_rate": 0.002109243697478992,
|
| 134 |
+
"loss": 0.8649,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4989495798319328,
|
| 139 |
+
"grad_norm": 3.658193588256836,
|
| 140 |
+
"learning_rate": 0.002004201680672269,
|
| 141 |
+
"loss": 0.821,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5252100840336135,
|
| 146 |
+
"grad_norm": 3.557441234588623,
|
| 147 |
+
"learning_rate": 0.0018991596638655462,
|
| 148 |
+
"loss": 0.8908,
|
| 149 |
+
"step": 1000
|
| 150 |
+
}
|
| 151 |
+
],
|
| 152 |
+
"logging_steps": 50,
|
| 153 |
+
"max_steps": 1904,
|
| 154 |
+
"num_input_tokens_seen": 0,
|
| 155 |
+
"num_train_epochs": 1,
|
| 156 |
+
"save_steps": 200,
|
| 157 |
+
"total_flos": 0.0,
|
| 158 |
+
"train_batch_size": 16,
|
| 159 |
+
"trial_name": null,
|
| 160 |
+
"trial_params": null
|
| 161 |
+
}
|
checkpoint-1200/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d022822088675c635566872739321fa17805d2814d46fc73c46bdd68e4449195
|
| 3 |
+
size 100495
|
checkpoint-1200/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-1200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0f9b8a351541c41f35a69dbfbfb51b6eec9be217dd5882cb4bf14c765086790
|
| 3 |
+
size 200278
|
checkpoint-1200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-1200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:155b2bf91a946a945b524da51bf0ba5adfae52ef39ae2a7730ac53083d8f3e56
|
| 3 |
+
size 1064
|
checkpoint-1200/trainer_state.json
ADDED
|
@@ -0,0 +1,189 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.6302521008403361,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.34138655462184875,
|
| 97 |
+
"grad_norm": 2.95237398147583,
|
| 98 |
+
"learning_rate": 0.002634453781512605,
|
| 99 |
+
"loss": 0.9322,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36764705882352944,
|
| 104 |
+
"grad_norm": 3.39207124710083,
|
| 105 |
+
"learning_rate": 0.0025294117647058825,
|
| 106 |
+
"loss": 0.939,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3939075630252101,
|
| 111 |
+
"grad_norm": 3.5582635402679443,
|
| 112 |
+
"learning_rate": 0.0024243697478991597,
|
| 113 |
+
"loss": 1.0108,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42016806722689076,
|
| 118 |
+
"grad_norm": 3.2852275371551514,
|
| 119 |
+
"learning_rate": 0.0023193277310924374,
|
| 120 |
+
"loss": 0.884,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.44642857142857145,
|
| 125 |
+
"grad_norm": 3.8662281036376953,
|
| 126 |
+
"learning_rate": 0.0022142857142857146,
|
| 127 |
+
"loss": 0.8889,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4726890756302521,
|
| 132 |
+
"grad_norm": 3.1788485050201416,
|
| 133 |
+
"learning_rate": 0.002109243697478992,
|
| 134 |
+
"loss": 0.8649,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4989495798319328,
|
| 139 |
+
"grad_norm": 3.658193588256836,
|
| 140 |
+
"learning_rate": 0.002004201680672269,
|
| 141 |
+
"loss": 0.821,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5252100840336135,
|
| 146 |
+
"grad_norm": 3.557441234588623,
|
| 147 |
+
"learning_rate": 0.0018991596638655462,
|
| 148 |
+
"loss": 0.8908,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5514705882352942,
|
| 153 |
+
"grad_norm": 3.6631717681884766,
|
| 154 |
+
"learning_rate": 0.0017941176470588236,
|
| 155 |
+
"loss": 0.8805,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5777310924369747,
|
| 160 |
+
"grad_norm": 3.9701390266418457,
|
| 161 |
+
"learning_rate": 0.001689075630252101,
|
| 162 |
+
"loss": 0.8352,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6039915966386554,
|
| 167 |
+
"grad_norm": 3.4215235710144043,
|
| 168 |
+
"learning_rate": 0.0015840336134453781,
|
| 169 |
+
"loss": 0.8129,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6302521008403361,
|
| 174 |
+
"grad_norm": 3.4287657737731934,
|
| 175 |
+
"learning_rate": 0.0014789915966386556,
|
| 176 |
+
"loss": 0.8299,
|
| 177 |
+
"step": 1200
|
| 178 |
+
}
|
| 179 |
+
],
|
| 180 |
+
"logging_steps": 50,
|
| 181 |
+
"max_steps": 1904,
|
| 182 |
+
"num_input_tokens_seen": 0,
|
| 183 |
+
"num_train_epochs": 1,
|
| 184 |
+
"save_steps": 200,
|
| 185 |
+
"total_flos": 0.0,
|
| 186 |
+
"train_batch_size": 16,
|
| 187 |
+
"trial_name": null,
|
| 188 |
+
"trial_params": null
|
| 189 |
+
}
|
checkpoint-1400/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9d59f7321d394348ba0ea481a0e2b120f3351e0f8c1bb2aed86b2f27c757f37
|
| 3 |
+
size 100495
|
checkpoint-1400/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-1400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0df335269a42c900a242f926478ed73e1fe66cac8988098137d7919132ce71a3
|
| 3 |
+
size 200278
|
checkpoint-1400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-1400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb5c427e88a59d16dcd6caad254ea07e88e8f0431507c809abfab80414f98959
|
| 3 |
+
size 1064
|
checkpoint-1400/trainer_state.json
ADDED
|
@@ -0,0 +1,217 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7352941176470589,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.34138655462184875,
|
| 97 |
+
"grad_norm": 2.95237398147583,
|
| 98 |
+
"learning_rate": 0.002634453781512605,
|
| 99 |
+
"loss": 0.9322,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36764705882352944,
|
| 104 |
+
"grad_norm": 3.39207124710083,
|
| 105 |
+
"learning_rate": 0.0025294117647058825,
|
| 106 |
+
"loss": 0.939,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3939075630252101,
|
| 111 |
+
"grad_norm": 3.5582635402679443,
|
| 112 |
+
"learning_rate": 0.0024243697478991597,
|
| 113 |
+
"loss": 1.0108,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42016806722689076,
|
| 118 |
+
"grad_norm": 3.2852275371551514,
|
| 119 |
+
"learning_rate": 0.0023193277310924374,
|
| 120 |
+
"loss": 0.884,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.44642857142857145,
|
| 125 |
+
"grad_norm": 3.8662281036376953,
|
| 126 |
+
"learning_rate": 0.0022142857142857146,
|
| 127 |
+
"loss": 0.8889,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4726890756302521,
|
| 132 |
+
"grad_norm": 3.1788485050201416,
|
| 133 |
+
"learning_rate": 0.002109243697478992,
|
| 134 |
+
"loss": 0.8649,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4989495798319328,
|
| 139 |
+
"grad_norm": 3.658193588256836,
|
| 140 |
+
"learning_rate": 0.002004201680672269,
|
| 141 |
+
"loss": 0.821,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5252100840336135,
|
| 146 |
+
"grad_norm": 3.557441234588623,
|
| 147 |
+
"learning_rate": 0.0018991596638655462,
|
| 148 |
+
"loss": 0.8908,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5514705882352942,
|
| 153 |
+
"grad_norm": 3.6631717681884766,
|
| 154 |
+
"learning_rate": 0.0017941176470588236,
|
| 155 |
+
"loss": 0.8805,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5777310924369747,
|
| 160 |
+
"grad_norm": 3.9701390266418457,
|
| 161 |
+
"learning_rate": 0.001689075630252101,
|
| 162 |
+
"loss": 0.8352,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6039915966386554,
|
| 167 |
+
"grad_norm": 3.4215235710144043,
|
| 168 |
+
"learning_rate": 0.0015840336134453781,
|
| 169 |
+
"loss": 0.8129,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6302521008403361,
|
| 174 |
+
"grad_norm": 3.4287657737731934,
|
| 175 |
+
"learning_rate": 0.0014789915966386556,
|
| 176 |
+
"loss": 0.8299,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6565126050420168,
|
| 181 |
+
"grad_norm": 3.0316977500915527,
|
| 182 |
+
"learning_rate": 0.0013739495798319329,
|
| 183 |
+
"loss": 0.8371,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6827731092436975,
|
| 188 |
+
"grad_norm": 3.983886957168579,
|
| 189 |
+
"learning_rate": 0.0012689075630252101,
|
| 190 |
+
"loss": 0.816,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7090336134453782,
|
| 195 |
+
"grad_norm": 4.039212226867676,
|
| 196 |
+
"learning_rate": 0.0011638655462184876,
|
| 197 |
+
"loss": 0.791,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.7352941176470589,
|
| 202 |
+
"grad_norm": 3.5772385597229004,
|
| 203 |
+
"learning_rate": 0.0010588235294117648,
|
| 204 |
+
"loss": 0.792,
|
| 205 |
+
"step": 1400
|
| 206 |
+
}
|
| 207 |
+
],
|
| 208 |
+
"logging_steps": 50,
|
| 209 |
+
"max_steps": 1904,
|
| 210 |
+
"num_input_tokens_seen": 0,
|
| 211 |
+
"num_train_epochs": 1,
|
| 212 |
+
"save_steps": 200,
|
| 213 |
+
"total_flos": 0.0,
|
| 214 |
+
"train_batch_size": 16,
|
| 215 |
+
"trial_name": null,
|
| 216 |
+
"trial_params": null
|
| 217 |
+
}
|
checkpoint-1600/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a73fe8be191cd4ea3f21ac28ffca7d6ac869c895602a768b55375bd77c820677
|
| 3 |
+
size 100495
|
checkpoint-1600/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-1600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d8aae9b1402beeec8e42154c2f9d9c8480a7c46e12ff56e87f8a7d8423b7073
|
| 3 |
+
size 200278
|
checkpoint-1600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-1600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab6140a19b1dfedc4c655ca2e90d999f38500820b42803d90f0bf7a2d8d1c4b7
|
| 3 |
+
size 1064
|
checkpoint-1600/trainer_state.json
ADDED
|
@@ -0,0 +1,245 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.8403361344537815,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.34138655462184875,
|
| 97 |
+
"grad_norm": 2.95237398147583,
|
| 98 |
+
"learning_rate": 0.002634453781512605,
|
| 99 |
+
"loss": 0.9322,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36764705882352944,
|
| 104 |
+
"grad_norm": 3.39207124710083,
|
| 105 |
+
"learning_rate": 0.0025294117647058825,
|
| 106 |
+
"loss": 0.939,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3939075630252101,
|
| 111 |
+
"grad_norm": 3.5582635402679443,
|
| 112 |
+
"learning_rate": 0.0024243697478991597,
|
| 113 |
+
"loss": 1.0108,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42016806722689076,
|
| 118 |
+
"grad_norm": 3.2852275371551514,
|
| 119 |
+
"learning_rate": 0.0023193277310924374,
|
| 120 |
+
"loss": 0.884,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.44642857142857145,
|
| 125 |
+
"grad_norm": 3.8662281036376953,
|
| 126 |
+
"learning_rate": 0.0022142857142857146,
|
| 127 |
+
"loss": 0.8889,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4726890756302521,
|
| 132 |
+
"grad_norm": 3.1788485050201416,
|
| 133 |
+
"learning_rate": 0.002109243697478992,
|
| 134 |
+
"loss": 0.8649,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4989495798319328,
|
| 139 |
+
"grad_norm": 3.658193588256836,
|
| 140 |
+
"learning_rate": 0.002004201680672269,
|
| 141 |
+
"loss": 0.821,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5252100840336135,
|
| 146 |
+
"grad_norm": 3.557441234588623,
|
| 147 |
+
"learning_rate": 0.0018991596638655462,
|
| 148 |
+
"loss": 0.8908,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5514705882352942,
|
| 153 |
+
"grad_norm": 3.6631717681884766,
|
| 154 |
+
"learning_rate": 0.0017941176470588236,
|
| 155 |
+
"loss": 0.8805,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5777310924369747,
|
| 160 |
+
"grad_norm": 3.9701390266418457,
|
| 161 |
+
"learning_rate": 0.001689075630252101,
|
| 162 |
+
"loss": 0.8352,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6039915966386554,
|
| 167 |
+
"grad_norm": 3.4215235710144043,
|
| 168 |
+
"learning_rate": 0.0015840336134453781,
|
| 169 |
+
"loss": 0.8129,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6302521008403361,
|
| 174 |
+
"grad_norm": 3.4287657737731934,
|
| 175 |
+
"learning_rate": 0.0014789915966386556,
|
| 176 |
+
"loss": 0.8299,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6565126050420168,
|
| 181 |
+
"grad_norm": 3.0316977500915527,
|
| 182 |
+
"learning_rate": 0.0013739495798319329,
|
| 183 |
+
"loss": 0.8371,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6827731092436975,
|
| 188 |
+
"grad_norm": 3.983886957168579,
|
| 189 |
+
"learning_rate": 0.0012689075630252101,
|
| 190 |
+
"loss": 0.816,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7090336134453782,
|
| 195 |
+
"grad_norm": 4.039212226867676,
|
| 196 |
+
"learning_rate": 0.0011638655462184876,
|
| 197 |
+
"loss": 0.791,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.7352941176470589,
|
| 202 |
+
"grad_norm": 3.5772385597229004,
|
| 203 |
+
"learning_rate": 0.0010588235294117648,
|
| 204 |
+
"loss": 0.792,
|
| 205 |
+
"step": 1400
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7615546218487395,
|
| 209 |
+
"grad_norm": 3.9995005130767822,
|
| 210 |
+
"learning_rate": 0.0009537815126050421,
|
| 211 |
+
"loss": 0.8089,
|
| 212 |
+
"step": 1450
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7878151260504201,
|
| 216 |
+
"grad_norm": 3.690575122833252,
|
| 217 |
+
"learning_rate": 0.0008487394957983193,
|
| 218 |
+
"loss": 0.733,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.8140756302521008,
|
| 223 |
+
"grad_norm": 3.8193883895874023,
|
| 224 |
+
"learning_rate": 0.0007436974789915967,
|
| 225 |
+
"loss": 0.7532,
|
| 226 |
+
"step": 1550
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8403361344537815,
|
| 230 |
+
"grad_norm": 3.2196240425109863,
|
| 231 |
+
"learning_rate": 0.0006386554621848739,
|
| 232 |
+
"loss": 0.7209,
|
| 233 |
+
"step": 1600
|
| 234 |
+
}
|
| 235 |
+
],
|
| 236 |
+
"logging_steps": 50,
|
| 237 |
+
"max_steps": 1904,
|
| 238 |
+
"num_input_tokens_seen": 0,
|
| 239 |
+
"num_train_epochs": 1,
|
| 240 |
+
"save_steps": 200,
|
| 241 |
+
"total_flos": 0.0,
|
| 242 |
+
"train_batch_size": 16,
|
| 243 |
+
"trial_name": null,
|
| 244 |
+
"trial_params": null
|
| 245 |
+
}
|
checkpoint-1800/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76b3b6a42298c52be79525ee7c3444b3f1c7371293fc055b47f005babdadb755
|
| 3 |
+
size 100495
|
checkpoint-1800/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0a43eed334c026afe4b13004a2f17a9f37ac2ee9ef6bfc60c227a418fde1783
|
| 3 |
+
size 200278
|
checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc12fc7a728a31f1f4ffb0fd0f29c08a37a9f60d15017c44f95681461c804ef4
|
| 3 |
+
size 1064
|
checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,273 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9453781512605042,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.34138655462184875,
|
| 97 |
+
"grad_norm": 2.95237398147583,
|
| 98 |
+
"learning_rate": 0.002634453781512605,
|
| 99 |
+
"loss": 0.9322,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36764705882352944,
|
| 104 |
+
"grad_norm": 3.39207124710083,
|
| 105 |
+
"learning_rate": 0.0025294117647058825,
|
| 106 |
+
"loss": 0.939,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3939075630252101,
|
| 111 |
+
"grad_norm": 3.5582635402679443,
|
| 112 |
+
"learning_rate": 0.0024243697478991597,
|
| 113 |
+
"loss": 1.0108,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42016806722689076,
|
| 118 |
+
"grad_norm": 3.2852275371551514,
|
| 119 |
+
"learning_rate": 0.0023193277310924374,
|
| 120 |
+
"loss": 0.884,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.44642857142857145,
|
| 125 |
+
"grad_norm": 3.8662281036376953,
|
| 126 |
+
"learning_rate": 0.0022142857142857146,
|
| 127 |
+
"loss": 0.8889,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4726890756302521,
|
| 132 |
+
"grad_norm": 3.1788485050201416,
|
| 133 |
+
"learning_rate": 0.002109243697478992,
|
| 134 |
+
"loss": 0.8649,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4989495798319328,
|
| 139 |
+
"grad_norm": 3.658193588256836,
|
| 140 |
+
"learning_rate": 0.002004201680672269,
|
| 141 |
+
"loss": 0.821,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5252100840336135,
|
| 146 |
+
"grad_norm": 3.557441234588623,
|
| 147 |
+
"learning_rate": 0.0018991596638655462,
|
| 148 |
+
"loss": 0.8908,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5514705882352942,
|
| 153 |
+
"grad_norm": 3.6631717681884766,
|
| 154 |
+
"learning_rate": 0.0017941176470588236,
|
| 155 |
+
"loss": 0.8805,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5777310924369747,
|
| 160 |
+
"grad_norm": 3.9701390266418457,
|
| 161 |
+
"learning_rate": 0.001689075630252101,
|
| 162 |
+
"loss": 0.8352,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6039915966386554,
|
| 167 |
+
"grad_norm": 3.4215235710144043,
|
| 168 |
+
"learning_rate": 0.0015840336134453781,
|
| 169 |
+
"loss": 0.8129,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6302521008403361,
|
| 174 |
+
"grad_norm": 3.4287657737731934,
|
| 175 |
+
"learning_rate": 0.0014789915966386556,
|
| 176 |
+
"loss": 0.8299,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6565126050420168,
|
| 181 |
+
"grad_norm": 3.0316977500915527,
|
| 182 |
+
"learning_rate": 0.0013739495798319329,
|
| 183 |
+
"loss": 0.8371,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6827731092436975,
|
| 188 |
+
"grad_norm": 3.983886957168579,
|
| 189 |
+
"learning_rate": 0.0012689075630252101,
|
| 190 |
+
"loss": 0.816,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7090336134453782,
|
| 195 |
+
"grad_norm": 4.039212226867676,
|
| 196 |
+
"learning_rate": 0.0011638655462184876,
|
| 197 |
+
"loss": 0.791,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.7352941176470589,
|
| 202 |
+
"grad_norm": 3.5772385597229004,
|
| 203 |
+
"learning_rate": 0.0010588235294117648,
|
| 204 |
+
"loss": 0.792,
|
| 205 |
+
"step": 1400
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7615546218487395,
|
| 209 |
+
"grad_norm": 3.9995005130767822,
|
| 210 |
+
"learning_rate": 0.0009537815126050421,
|
| 211 |
+
"loss": 0.8089,
|
| 212 |
+
"step": 1450
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7878151260504201,
|
| 216 |
+
"grad_norm": 3.690575122833252,
|
| 217 |
+
"learning_rate": 0.0008487394957983193,
|
| 218 |
+
"loss": 0.733,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.8140756302521008,
|
| 223 |
+
"grad_norm": 3.8193883895874023,
|
| 224 |
+
"learning_rate": 0.0007436974789915967,
|
| 225 |
+
"loss": 0.7532,
|
| 226 |
+
"step": 1550
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8403361344537815,
|
| 230 |
+
"grad_norm": 3.2196240425109863,
|
| 231 |
+
"learning_rate": 0.0006386554621848739,
|
| 232 |
+
"loss": 0.7209,
|
| 233 |
+
"step": 1600
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8665966386554622,
|
| 237 |
+
"grad_norm": 3.2536444664001465,
|
| 238 |
+
"learning_rate": 0.0005336134453781513,
|
| 239 |
+
"loss": 0.7259,
|
| 240 |
+
"step": 1650
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8928571428571429,
|
| 244 |
+
"grad_norm": 4.125887870788574,
|
| 245 |
+
"learning_rate": 0.00042857142857142855,
|
| 246 |
+
"loss": 0.7055,
|
| 247 |
+
"step": 1700
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.9191176470588235,
|
| 251 |
+
"grad_norm": 3.512568950653076,
|
| 252 |
+
"learning_rate": 0.0003235294117647059,
|
| 253 |
+
"loss": 0.7074,
|
| 254 |
+
"step": 1750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.9453781512605042,
|
| 258 |
+
"grad_norm": 4.172778129577637,
|
| 259 |
+
"learning_rate": 0.00021848739495798319,
|
| 260 |
+
"loss": 0.7267,
|
| 261 |
+
"step": 1800
|
| 262 |
+
}
|
| 263 |
+
],
|
| 264 |
+
"logging_steps": 50,
|
| 265 |
+
"max_steps": 1904,
|
| 266 |
+
"num_input_tokens_seen": 0,
|
| 267 |
+
"num_train_epochs": 1,
|
| 268 |
+
"save_steps": 200,
|
| 269 |
+
"total_flos": 0.0,
|
| 270 |
+
"train_batch_size": 16,
|
| 271 |
+
"trial_name": null,
|
| 272 |
+
"trial_params": null
|
| 273 |
+
}
|
checkpoint-200/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49fa9f7499bac5e9d7c91100eb58a3c512329b964ae841f2661e4e25b7a4f606
|
| 3 |
+
size 100495
|
checkpoint-200/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a8ff8877261f03e9fe268f140f8041b0f4b452d6791f0cce3b5354200a259d1
|
| 3 |
+
size 200278
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2399d10be099bb6a422c301f42308b5349474a2d05adc1340e0aa507c4cfd0da
|
| 3 |
+
size 1064
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.10504201680672269,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
}
|
| 39 |
+
],
|
| 40 |
+
"logging_steps": 50,
|
| 41 |
+
"max_steps": 1904,
|
| 42 |
+
"num_input_tokens_seen": 0,
|
| 43 |
+
"num_train_epochs": 1,
|
| 44 |
+
"save_steps": 200,
|
| 45 |
+
"total_flos": 0.0,
|
| 46 |
+
"train_batch_size": 16,
|
| 47 |
+
"trial_name": null,
|
| 48 |
+
"trial_params": null
|
| 49 |
+
}
|
checkpoint-400/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61d649e22fd8e6fa9f88fe4bd122fb8b5acc476342224ddf06fa03745abe1b7c
|
| 3 |
+
size 100495
|
checkpoint-400/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7497b92e2a3e2a407fee6878c225d22799bfe2bf8fc7a5825bcf6424ed93b1a
|
| 3 |
+
size 200278
|
checkpoint-400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49d756189cb2fe1825785ee87db97cd99909f07088a98a2d99e01491ef230e3a
|
| 3 |
+
size 1064
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.21008403361344538,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
}
|
| 67 |
+
],
|
| 68 |
+
"logging_steps": 50,
|
| 69 |
+
"max_steps": 1904,
|
| 70 |
+
"num_input_tokens_seen": 0,
|
| 71 |
+
"num_train_epochs": 1,
|
| 72 |
+
"save_steps": 200,
|
| 73 |
+
"total_flos": 0.0,
|
| 74 |
+
"train_batch_size": 16,
|
| 75 |
+
"trial_name": null,
|
| 76 |
+
"trial_params": null
|
| 77 |
+
}
|
checkpoint-600/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13bbbc56ad503bb03afe2440e21c4c4085b142e9f1996b6cc0ea1b785c78154f
|
| 3 |
+
size 100495
|
checkpoint-600/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|
checkpoint-600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0a31666e611550d061987c3bcc103234265a1229595a4dc17a4d94ac13c534ef
|
| 3 |
+
size 200278
|
checkpoint-600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0557676e839d64593261cf638b5495ef95e52a7a07cbe64044f2838d0147dbbf
|
| 3 |
+
size 1064
|
checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.31512605042016806,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026260504201680673,
|
| 13 |
+
"grad_norm": 2.606417417526245,
|
| 14 |
+
"learning_rate": 0.0038949579831932773,
|
| 15 |
+
"loss": 1.2269,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052521008403361345,
|
| 20 |
+
"grad_norm": 2.463599920272827,
|
| 21 |
+
"learning_rate": 0.0037899159663865546,
|
| 22 |
+
"loss": 1.1055,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07878151260504201,
|
| 27 |
+
"grad_norm": 3.1962060928344727,
|
| 28 |
+
"learning_rate": 0.003684873949579832,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10504201680672269,
|
| 34 |
+
"grad_norm": 3.148374557495117,
|
| 35 |
+
"learning_rate": 0.0035798319327731095,
|
| 36 |
+
"loss": 1.1056,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13130252100840337,
|
| 41 |
+
"grad_norm": 2.5041284561157227,
|
| 42 |
+
"learning_rate": 0.0034747899159663868,
|
| 43 |
+
"loss": 1.0834,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15756302521008403,
|
| 48 |
+
"grad_norm": 2.5208585262298584,
|
| 49 |
+
"learning_rate": 0.003369747899159664,
|
| 50 |
+
"loss": 1.1076,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18382352941176472,
|
| 55 |
+
"grad_norm": 2.758544445037842,
|
| 56 |
+
"learning_rate": 0.0032647058823529413,
|
| 57 |
+
"loss": 1.0285,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.21008403361344538,
|
| 62 |
+
"grad_norm": 3.854949951171875,
|
| 63 |
+
"learning_rate": 0.0031596638655462185,
|
| 64 |
+
"loss": 1.0204,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23634453781512604,
|
| 69 |
+
"grad_norm": 3.600606918334961,
|
| 70 |
+
"learning_rate": 0.0030546218487394958,
|
| 71 |
+
"loss": 1.0311,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.26260504201680673,
|
| 76 |
+
"grad_norm": 3.2583634853363037,
|
| 77 |
+
"learning_rate": 0.0029495798319327735,
|
| 78 |
+
"loss": 0.9868,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28886554621848737,
|
| 83 |
+
"grad_norm": 3.2210769653320312,
|
| 84 |
+
"learning_rate": 0.0028445378151260507,
|
| 85 |
+
"loss": 1.0129,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31512605042016806,
|
| 90 |
+
"grad_norm": 3.139937162399292,
|
| 91 |
+
"learning_rate": 0.002739495798319328,
|
| 92 |
+
"loss": 1.0096,
|
| 93 |
+
"step": 600
|
| 94 |
+
}
|
| 95 |
+
],
|
| 96 |
+
"logging_steps": 50,
|
| 97 |
+
"max_steps": 1904,
|
| 98 |
+
"num_input_tokens_seen": 0,
|
| 99 |
+
"num_train_epochs": 1,
|
| 100 |
+
"save_steps": 200,
|
| 101 |
+
"total_flos": 0.0,
|
| 102 |
+
"train_batch_size": 16,
|
| 103 |
+
"trial_name": null,
|
| 104 |
+
"trial_params": null
|
| 105 |
+
}
|
checkpoint-800/intervenable_model/intkey_layer.15.comp.block_output.unit.pos.nunit.1#0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f42a834053c2c2e13fc466cde620ef51b1f862f057e4a4630c81dbeecb2fff9
|
| 3 |
+
size 100495
|
checkpoint-800/intervenable_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
| 3 |
+
size 888
|