Upload folder using huggingface_hub

multi_head_checkpoints/hedging_head/ckpt_10000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a43d30ab3e87d8e7dc70c62da5ca5b49f54e272713969e87c5f3a742e485871d
+size 24186

multi_head_checkpoints/hedging_head/ckpt_1200/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe1250ad52f4442762d815d82d063dd70aae4d3ad37a2b593ded5530633eadc2
+size 24186

multi_head_checkpoints/hedging_head/ckpt_1600/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:410b73b80fb3f9929b6996638b064e9f80aed508b0b1fb98ee4cb10bef150093
+size 24186

multi_head_checkpoints/hedging_head/ckpt_2000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40ba77182937063ffe3b0c7092cb20e50832aa18431bdea18de2ee53600d26b2
+size 24186

multi_head_checkpoints/hedging_head/ckpt_3000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbe7ff81d664c8c198c36beaefb36828bb18abc9b2407ec864a76c14293998a8
+size 24186

multi_head_checkpoints/hedging_head/ckpt_400/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d6126ed08f5828c5faed656b8ac1b8fc020794d68c59714fcf03125bd817f88
+size 24186

multi_head_checkpoints/hedging_head/ckpt_4000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff4e21e2f77fb0acfa0af420fc6f88508c9147bcaf2185568431656ba70ce19
+size 24186

multi_head_checkpoints/hedging_head/ckpt_5000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b98ca325d455e8d82762f41d69e193e40848d42cf447a02503bf377b5684aee
+size 24186

multi_head_checkpoints/hedging_head/ckpt_6000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1bc8cbe73a8f55fa9117ff49ce7d53198234fe8974849c558dee1464209c211
+size 24186

multi_head_checkpoints/hedging_head/ckpt_7000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08ec56ead6368a5ed9ebc2aaabb57db9569d7645073fa53a7b1801a83bfd95ae
+size 24186

multi_head_checkpoints/hedging_head/ckpt_800/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:498b1960a6fbf98c2ec396be137f35ed39eb81c4e0147b699d41dae04dc5d6fa
+size 24186

multi_head_checkpoints/hedging_head/ckpt_8000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:688c95d351fd4ef34f3d78917edc530554978724e17024e40f0daeb429d1f1a2
+size 24186

multi_head_checkpoints/hedging_head/ckpt_9000/hedging_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a0faf4accac64797b552fc5b5a4bb401bf05d24ae24e00f3f2f6784598dfde3
+size 24186

multi_head_checkpoints/hedging_head/results.json

@@ -0,0 +1,52 @@
+[
+  {
+    "step": 400,
+    "accuracy": 0.4442897439002991,
+    "precision": 0.03143163271296894,
+    "recall": 0.8222307544634286,
+    "f1": 0.06054865592727941,
+    "pos_risk": 0.520409107208252,
+    "neg_risk": 0.49916452169418335,
+    "separation": 1.0425602874217976
+  },
+  {
+    "step": 800,
+    "accuracy": 0.7166696190834045,
+    "precision": 0.05191828314374946,
+    "recall": 0.6957189479746593,
+    "f1": 0.09662582821186226,
+    "pos_risk": 0.5107523798942566,
+    "neg_risk": 0.4790365397930145,
+    "separation": 1.0662075592708358
+  },
+  {
+    "step": 1200,
+    "accuracy": 0.8910630941390991,
+    "precision": 0.10129681343483417,
+    "recall": 0.5083509310808216,
+    "f1": 0.16893141945773524,
+    "pos_risk": 0.493786096572876,
+    "neg_risk": 0.45646682381629944,
+    "separation": 1.081756813002461
+  },
+  {
+    "step": 1600,
+    "accuracy": 0.8538420796394348,
+    "precision": 0.07995142477901099,
+    "recall": 0.5434824342484162,
+    "f1": 0.13939632675168645,
+    "pos_risk": 0.4988616406917572,
+    "neg_risk": 0.45187142491340637,
+    "separation": 1.1039902352474615
+  },
+  {
+    "step": 2000,
+    "accuracy": 0.8356873393058777,
+    "precision": 0.07362101313320825,
+    "recall": 0.5649836820886927,
+    "f1": 0.13026735127478753,
+    "pos_risk": 0.5041213035583496,
+    "neg_risk": 0.44911104440689087,
+    "separation": 1.122486988099139
+  }
+]

multi_head_checkpoints/hedging_head/results_continued.json

@@ -0,0 +1,82 @@
+[
+  {
+    "step": 3000,
+    "accuracy": 0.7798140048980713,
+    "precision": 0.06394045212277155,
+    "recall": 0.6678825110385871,
+    "f1": 0.11670776094869087,
+    "pos_risk": 0.524723470211029,
+    "neg_risk": 0.4489431381225586,
+    "separation": 1.1687971719656465
+  },
+  {
+    "step": 4000,
+    "accuracy": 0.8607996106147766,
+    "precision": 0.08731814842027921,
+    "recall": 0.5703589940487618,
+    "f1": 0.15145027272263856,
+    "pos_risk": 0.5151649713516235,
+    "neg_risk": 0.4292229413986206,
+    "separation": 1.2002270187911235
+  },
+  {
+    "step": 5000,
+    "accuracy": 0.8619410991668701,
+    "precision": 0.09372991293168936,
+    "recall": 0.6158571702822039,
+    "f1": 0.1626981108152656,
+    "pos_risk": 0.5229318737983704,
+    "neg_risk": 0.42363420128822327,
+    "separation": 1.234394843967258
+  },
+  {
+    "step": 6000,
+    "accuracy": 0.873987078666687,
+    "precision": 0.10097000352146493,
+    "recall": 0.6054904972163563,
+    "f1": 0.17307797837897163,
+    "pos_risk": 0.5275717377662659,
+    "neg_risk": 0.4137572944164276,
+    "separation": 1.2750753760374536
+  },
+  {
+    "step": 7000,
+    "accuracy": 0.9015830159187317,
+    "precision": 0.12172369670202667,
+    "recall": 0.5661355346515646,
+    "f1": 0.20036689767631471,
+    "pos_risk": 0.521373987197876,
+    "neg_risk": 0.3951682150363922,
+    "separation": 1.319372275803764
+  },
+  {
+    "step": 8000,
+    "accuracy": 0.8688943982124329,
+    "precision": 0.09942703067071115,
+    "recall": 0.6229602610865809,
+    "f1": 0.1714844369286054,
+    "pos_risk": 0.5516535639762878,
+    "neg_risk": 0.40235474705696106,
+    "separation": 1.3710626456165327
+  },
+  {
+    "step": 9000,
+    "accuracy": 0.8865934014320374,
+    "precision": 0.11200424929178471,
+    "recall": 0.6072182760606643,
+    "f1": 0.18912374062004847,
+    "pos_risk": 0.5500459671020508,
+    "neg_risk": 0.3843628168106079,
+    "separation": 1.4310592571525516
+  },
+  {
+    "step": 10000,
+    "accuracy": 0.8839634656906128,
+    "precision": 0.11537280327589149,
+    "recall": 0.6490689191783452,
+    "f1": 0.19592049603059628,
+    "pos_risk": 0.5594488978385925,
+    "neg_risk": 0.37506988644599915,
+    "separation": 1.491585749897678
+  }
+]

multi_head_checkpoints/multi_head_v2_complete.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80ca660754225782c07dbbbdb7e192e6f1b96ee805b451f5e6f8aebfceb9fa67
+size 8496958

multi_head_checkpoints/sycophancy_head/ckpt_1200/sycophancy_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0572e9817f8c4bd9e7e7e46d29c4561104e44f13880bdb63bfee55a6f3460c7d
+size 24216

multi_head_checkpoints/sycophancy_head/ckpt_1600/sycophancy_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e17b77b030a879f8e552edcb4db77e6df42e18f79925fa17c82f1e442693468
+size 24216

multi_head_checkpoints/sycophancy_head/ckpt_2000/sycophancy_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a410f16a20edcf2d1b5609c74e39bccae4d7ed0c7007b0eb15a39db984ba98e6
+size 24216

multi_head_checkpoints/sycophancy_head/ckpt_400/sycophancy_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d0c9ed5fc013d6c6403ff2c35d309abce93e5095647d8a05d8aff3d6dcac436
+size 24216

multi_head_checkpoints/sycophancy_head/ckpt_800/sycophancy_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc206bd212d68d5d587d18c146fb850dade23a88fba3c911d4563a3dec9b8881
+size 24216

multi_head_checkpoints/sycophancy_head/results.json

@@ -0,0 +1,52 @@
+[
+  {
+    "step": 400,
+    "accuracy": 0.35452115535736084,
+    "precision": 0.019912276473707042,
+    "recall": 0.5976851851851852,
+    "f1": 0.038540549113265106,
+    "pos_risk": 0.5032691359519958,
+    "neg_risk": 0.5036115050315857,
+    "separation": 0.9993201722435464
+  },
+  {
+    "step": 800,
+    "accuracy": 0.9783545732498169,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.46720242500305176,
+    "neg_risk": 0.46737194061279297,
+    "separation": 0.9996373004132021
+  },
+  {
+    "step": 1200,
+    "accuracy": 0.9783545732498169,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.43459680676460266,
+    "neg_risk": 0.4344600737094879,
+    "separation": 1.0003147194952744
+  },
+  {
+    "step": 1600,
+    "accuracy": 0.9783545732498169,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.4037657082080841,
+    "neg_risk": 0.40317171812057495,
+    "separation": 1.001473293043168
+  },
+  {
+    "step": 2000,
+    "accuracy": 0.9783545732498169,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.3766477108001709,
+    "neg_risk": 0.37547680735588074,
+    "separation": 1.0031184441258454
+  }
+]

multi_head_checkpoints/verbosity_head/ckpt_10000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba1118f564de6f41db58f48a44141cea2800a490e7b9f9646414c713af49dadb
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_1200/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b27546560854302d0d7034bc426e046e07a25bcac94e1e17702c948c3febf9bd
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_1600/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff23a7c3f43eaffe3741b0e1c696916c2b5dd567cc88e17c46ec625dace157fd
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_2000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e274dcc9bbb41a6ba7cdaaedd46adf022a9091bbf19fc6b6717a96fb54d0d849
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_3000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a7180cef50b2b84977e38061facd64280b9c227c936d0d72def1067d9f64b6d
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_400/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea379d75845ccfa19a68c2ed023f64c27f03e7740dbb565ecfa558cbacdbc35
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_4000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97ea4df585f6382d9cc0105c7c94b8a84fcd2348f4af4f29168a480e4fa89866
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_5000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fa4093499e35c4d573975206d96ef17f405e8218d9de69edba9a4fe7fff294a
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_6000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7e7974055ec42b5f2d76b5363936294ef61bb3f03172fa2e478ccb678568517
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_7000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9623f4b7a8e4666b67e51782eec6c7fdf0522502b78a575a44d575a3acb5b2b8
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_800/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09ac72679591c7bb8b50f40c5283fab7558045363e69f9a4f879671195b26be1
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_8000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3d212aa8be993b7e1de78de8aa2137538096c3cb0398863973525aab7f66d92
+size 24206

multi_head_checkpoints/verbosity_head/ckpt_9000/verbosity_head.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c8d969872da71ba1bad5d0161587b8424c594d3ee38bf01340c30d9b9419992
+size 24206

multi_head_checkpoints/verbosity_head/results.json

@@ -0,0 +1,52 @@
+[
+  {
+    "step": 400,
+    "accuracy": 0.9765997529029846,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.4664677381515503,
+    "neg_risk": 0.45480582118034363,
+    "separation": 1.025641529699292
+  },
+  {
+    "step": 800,
+    "accuracy": 0.9766002893447876,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.43757620453834534,
+    "neg_risk": 0.40636545419692993,
+    "separation": 1.0768046348897815
+  },
+  {
+    "step": 1200,
+    "accuracy": 0.9766002893447876,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.3934943377971649,
+    "neg_risk": 0.3480086922645569,
+    "separation": 1.1307026133072267
+  },
+  {
+    "step": 1600,
+    "accuracy": 0.9766002893447876,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.39113855361938477,
+    "neg_risk": 0.32403916120529175,
+    "separation": 1.2070718618222285
+  },
+  {
+    "step": 2000,
+    "accuracy": 0.9765997529029846,
+    "precision": 0.0,
+    "recall": 0.0,
+    "f1": 0.0,
+    "pos_risk": 0.3914262056350708,
+    "neg_risk": 0.3029381334781647,
+    "separation": 1.2920994829569195
+  }
+]

multi_head_checkpoints/verbosity_head/results_continued.json

@@ -0,0 +1,82 @@
+[
+  {
+    "step": 3000,
+    "accuracy": 0.9691389203071594,
+    "precision": 0.1369258846192842,
+    "recall": 0.06012644138729353,
+    "f1": 0.08356020294518005,
+    "pos_risk": 0.4014969766139984,
+    "neg_risk": 0.26297321915626526,
+    "separation": 1.52675994119165
+  },
+  {
+    "step": 4000,
+    "accuracy": 0.9612593054771423,
+    "precision": 0.11997728973650933,
+    "recall": 0.10349049463514537,
+    "f1": 0.11112571858828028,
+    "pos_risk": 0.3859938085079193,
+    "neg_risk": 0.21928799152374268,
+    "separation": 1.7602140720328825
+  },
+  {
+    "step": 5000,
+    "accuracy": 0.8646724820137024,
+    "precision": 0.0876178851490621,
+    "recall": 0.5081474555896888,
+    "f1": 0.14946423485272597,
+    "pos_risk": 0.45228344202041626,
+    "neg_risk": 0.2378995418548584,
+    "separation": 1.9011530602120816
+  },
+  {
+    "step": 6000,
+    "accuracy": 0.86240553855896,
+    "precision": 0.08743681522381432,
+    "recall": 0.5171408218690174,
+    "f1": 0.1495825968816301,
+    "pos_risk": 0.45121535658836365,
+    "neg_risk": 0.22329428791999817,
+    "separation": 2.0207205513023467
+  },
+  {
+    "step": 7000,
+    "accuracy": 0.8883561491966248,
+    "precision": 0.09316823884267353,
+    "recall": 0.4318151462535061,
+    "f1": 0.1532675426467451,
+    "pos_risk": 0.43008705973625183,
+    "neg_risk": 0.20528849959373474,
+    "separation": 2.0950372796693078
+  },
+  {
+    "step": 8000,
+    "accuracy": 0.8657205700874329,
+    "precision": 0.08880243245644875,
+    "recall": 0.5116646631939806,
+    "f1": 0.15133907260785828,
+    "pos_risk": 0.44835221767425537,
+    "neg_risk": 0.21308068931102753,
+    "separation": 2.1041428912397078
+  },
+  {
+    "step": 9000,
+    "accuracy": 0.853208601474762,
+    "precision": 0.08620888430834804,
+    "recall": 0.5493076888829527,
+    "f1": 0.1490290104089601,
+    "pos_risk": 0.45867228507995605,
+    "neg_risk": 0.21553963422775269,
+    "separation": 2.128018295675932
+  },
+  {
+    "step": 10000,
+    "accuracy": 0.8786846399307251,
+    "precision": 0.09251913030283324,
+    "recall": 0.4750456346556253,
+    "f1": 0.15487504399859206,
+    "pos_risk": 0.43932676315307617,
+    "neg_risk": 0.20558473467826843,
+    "separation": 2.1369619871854995
+  }
+]