Upload batch 1/100 (1000 files)

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f98f5052f78fec2ea15d9ead72ccb86a362ae54a024b20fa907c72d4122456a8
+size 2338457

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train": {
+    "0.05": 0.4071574074074074,
+    "mse": 0.015832414623744705,
+    "corr_pred_acc": 0.7864653605696242,
+    "corr_pred_cons": 0.4823919557947406,
+    "corr_cons_acc": 0.44309195382792343
+  },
+  "test": {
+    "0.05": 0.27375,
+    "mse": 0.029905065998778423,
+    "corr_pred_acc": 0.8314245229678434,
+    "corr_pred_cons": 0.8463382925414079,
+    "corr_cons_acc": 0.7633962506714751
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/math4_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ceaa6e67ca99d8346983dc176ef2bf7d3d837c79510579771093f0a6d28a3d9
+size 2333139

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/math4_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train": {
+    "0.05": 0.2975833333333333,
+    "mse": 0.06448148359519615,
+    "corr_pred_acc": 0.6678211771825778,
+    "corr_pred_cons": 0.5261592288337493,
+    "corr_cons_acc": 0.43775535550754996
+  },
+  "test": {
+    "0.05": 0.385,
+    "mse": 0.09341378532699074,
+    "corr_pred_acc": 0.603798890440722,
+    "corr_pred_cons": 0.7056754892420636,
+    "corr_cons_acc": 0.47175924918377005
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/medmcqa4_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c777af62e842b3044031b9080d56e347af03f91ca07649be4ad6e092a20d29c7
+size 2355237

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/medmcqa4_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train": {
+    "0.05": 0.16308609929861378,
+    "mse": 0.19311613863795818,
+    "corr_pred_acc": -0.008417739108695064,
+    "corr_pred_cons": 0.047386593638797686,
+    "corr_cons_acc": 0.3503191789280086
+  },
+  "test": {
+    "0.05": 0.17998845170337374,
+    "mse": 0.1779757197909025,
+    "corr_pred_acc": 0.00850936784460274,
+    "corr_pred_cons": 0.053754290062841284,
+    "corr_cons_acc": 0.33451280078254053
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/mmlupro10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26ac3b7842b62714a562ed2cbae3f05dd5663776dd62bcb77833cc11af1928f7
+size 218411

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/mmlupro10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train": {
+    "0.05": 0.2018961013833801,
+    "mse": 0.08535788942048819,
+    "corr_pred_acc": 0.012134102954893481,
+    "corr_pred_cons": -0.015998044539872077,
+    "corr_cons_acc": 0.3183910424921583
+  },
+  "test": {
+    "0.05": 0.17054908485856904,
+    "mse": 0.08478638330750508,
+    "corr_pred_acc": 0.0028937518953936767,
+    "corr_pred_cons": -0.06052726425409538,
+    "corr_cons_acc": 0.24217480944773512
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/mmlupro4_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecd3bf9dffcaa75690b6ab1812cac12138b21ca7cee152fd5c8c73727ccb95a9
+size 225701

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/evals/mmlupro4_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train": {
+    "0.05": 0.1831539541413196,
+    "mse": 0.17689470947584185,
+    "corr_pred_acc": -0.004735940227024761,
+    "corr_pred_cons": 0.008778995197567675,
+    "corr_cons_acc": 0.34949584200504086
+  },
+  "test": {
+    "0.05": 0.17117834394904458,
+    "mse": 0.16639991618988442,
+    "corr_pred_acc": -0.013678573101031254,
+    "corr_pred_cons": -0.023403762021411395,
+    "corr_cons_acc": 0.2082317949129946
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Llama-2-7b-chat_1_5e-05_0.01_8_16_0.1_probe-mid_08a33b98/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 32,
+  "n_layers_used": 16,
+  "hidden_size": 4096,
+  "model": "Llama-2-7b-chat",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-0/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c36c2c139454b85f83a6b7634cc7f66e6a3c63c22f639dcea76203fa82c2dda
+size 38681

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-0/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.047,
+    "mse": 0.4318077130098859,
+    "corr_pred_acc": 0.13520083758369467,
+    "corr_pred_cons": 0.10312126593748887,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.026,
+    "mse": 0.4475619079643741,
+    "corr_pred_acc": -0.08176350363824465,
+    "corr_pred_cons": -0.13798663835994307,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-0/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1005/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e24445e06fd422da5de220489bd5292b10f3fa7e7708152198a20655473563d1
+size 39321

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1005/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.074,
+    "mse": 0.11601532052432961,
+    "corr_pred_acc": 0.47587347480483644,
+    "corr_pred_cons": 0.3745852156101831,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.021,
+    "mse": 0.23174904289117826,
+    "corr_pred_acc": 0.4084798902659725,
+    "corr_pred_cons": 0.3126782157994644,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1005/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1072/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb70c2897a6ab5ce00e093f8d9c0e3fc0236c065c8a7e79a8b0d3e6c1917a53f
+size 39641

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1072/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.315,
+    "mse": 0.13517623116881947,
+    "corr_pred_acc": 0.4513784696443651,
+    "corr_pred_cons": 0.3664022490417258,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.332,
+    "mse": 0.16533499033782115,
+    "corr_pred_acc": 0.6361833182833748,
+    "corr_pred_cons": 0.4886728109268638,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1072/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1139/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a434d684058df987c529725e1d0c9419ffee58e27e75286d31e8be9e17a9825
+size 40025

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1139/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.065,
+    "mse": 0.11435365758057574,
+    "corr_pred_acc": 0.4196706858007531,
+    "corr_pred_cons": 0.30996524621574495,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.025,
+    "mse": 0.12273837224805777,
+    "corr_pred_acc": 0.5759579408142944,
+    "corr_pred_cons": 0.4193777495540827,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1139/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1206/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4cd73757afacc9ef2e94e8c5593f8d2212a6d4a8560b5eaa1656f9712021c17
+size 39705

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1206/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.061,
+    "mse": 0.1317964805358151,
+    "corr_pred_acc": 0.4082594562807137,
+    "corr_pred_cons": 0.3045582724812529,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.023,
+    "mse": 0.15724708221889666,
+    "corr_pred_acc": 0.4879981643641315,
+    "corr_pred_cons": 0.35214181190183,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1206/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1273/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61b077facc1aab311e1e98ccfb9678d635a3a6996a5c18d38860a5417dc7e271
+size 40025

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1273/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.353,
+    "mse": 0.14104830962235107,
+    "corr_pred_acc": 0.4549593657246854,
+    "corr_pred_cons": 0.35688186454107007,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.219,
+    "mse": 0.1721239071194548,
+    "corr_pred_acc": 0.5692263949497596,
+    "corr_pred_cons": 0.43629719119808014,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1273/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-134/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84b979eb1c2f8b9b5ecd37004371702bd688bd472d85928b1cc1088d2af7d633
+size 40729

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-134/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.043,
+    "mse": 0.12129868113433737,
+    "corr_pred_acc": 0.2576174988754247,
+    "corr_pred_cons": 0.16567774583131467,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.018,
+    "mse": 0.16170841795254706,
+    "corr_pred_acc": 0.2184891331747237,
+    "corr_pred_cons": 0.0698552717135327,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-134/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1340/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0340881097de39f01a5a0acfe5dbc036f649ce963ddc2b232a0ff3cdd2207aab
+size 39705

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1340/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.151,
+    "mse": 0.09828184091669206,
+    "corr_pred_acc": 0.49089877718214736,
+    "corr_pred_cons": 0.37699261760306485,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.029,
+    "mse": 0.1177501744904237,
+    "corr_pred_acc": 0.5630801672339811,
+    "corr_pred_cons": 0.42181336468219416,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1340/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1407/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86cf7c067586835fcf1ebb4cecc17d0f05c404e94ddd0020d976ee92afee9c65
+size 39321

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1407/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.061,
+    "mse": 0.17171390360904912,
+    "corr_pred_acc": 0.4491041124645969,
+    "corr_pred_cons": 0.33935563199336655,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.02,
+    "mse": 0.22520319899770955,
+    "corr_pred_acc": 0.500466484094762,
+    "corr_pred_cons": 0.35966813893492633,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1407/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1474/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cf4056285ffe134c53a98fb505dc70025bfa7a1a52c9c4bb8721c90be353c13
+size 39833

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1474/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.287,
+    "mse": 0.11670821748140475,
+    "corr_pred_acc": 0.49702887571696747,
+    "corr_pred_cons": 0.3895138894224595,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.225,
+    "mse": 0.1471094190504328,
+    "corr_pred_acc": 0.5868491215574767,
+    "corr_pred_cons": 0.4442161527501132,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1474/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1541/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fd98e885508e7d8d72b6efa5f395a0b5c124ef7a5d0b77d18312c935dd13033
+size 39705

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1541/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.278,
+    "mse": 0.1080773862052532,
+    "corr_pred_acc": 0.5041571590536615,
+    "corr_pred_cons": 0.3977274372156909,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.318,
+    "mse": 0.13658199544285676,
+    "corr_pred_acc": 0.6231714966048513,
+    "corr_pred_cons": 0.4750931964853024,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1541/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1608/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52d152d43d1d07bf56f30899aa5cc304893df16c1902c12e1ee631c696f3a4f2
+size 39641

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1608/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.07,
+    "mse": 0.12976799625811658,
+    "corr_pred_acc": 0.4607898733843373,
+    "corr_pred_cons": 0.3647214561377324,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.066,
+    "mse": 0.13941412589589056,
+    "corr_pred_acc": 0.5597032922810294,
+    "corr_pred_cons": 0.41965246789998856,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1608/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1675/evals/math10_categoriesbalanced.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2905fe58c0fb00125e530e97de2f222fc3816d724ee78d03b7bc8d26bc039494
+size 39769

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1675/evals/math10_categoriesbalanced_metrics.json

@@ -0,0 +1,16 @@
+{
+  "train_subset": {
+    "0.05": 0.238,
+    "mse": 0.10044774800527258,
+    "corr_pred_acc": 0.49369640261827963,
+    "corr_pred_cons": 0.4013204375728759,
+    "corr_cons_acc": 0.44940073047931717
+  },
+  "test": {
+    "0.05": 0.109,
+    "mse": 0.11009277139404725,
+    "corr_pred_acc": 0.6317166783549244,
+    "corr_pred_cons": 0.47839567299721336,
+    "corr_cons_acc": 0.5433606127799882
+  }
+}

models/default/unsloth_full/math10_categoriesbalanced_Qwen2.5-7B-Instruct_1_5e-05_0.01_8_16_0.1_probe-mid_9d45e9ca/checkpoint-1675/probe_meta.json

@@ -0,0 +1,9 @@
+{
+  "probe": "mid",
+  "n_layers_full": 28,
+  "n_layers_used": 14,
+  "hidden_size": 3584,
+  "model": "Qwen2.5-7B-Instruct",
+  "dataset": "math10_categoriesbalanced",
+  "lora_rank": 16
+}