Upload mini-beatrix-3-la2-6-12-20-c249k/reports/lookup_arm/arm_end_step253674.json with huggingface_hub

mini-beatrix-3-la2-6-12-20-c249k/reports/lookup_arm/arm_end_step253674.json

@@ -0,0 +1 @@
+{"tag": "arm_end", "step": 253674, "tokens": 66499117056, "craft": "mini-beatrix-3-la2-6-12-20-c249k", "sites": [2, 6, 12, 20], "arm_params": ["lookup_arms.", "lookup_norms."], "init_from": {"source": "AbstractPhil/alephllm-mini-beatrix-training/mini-beatrix-3-la2-6-12-20/checkpoints/step_00249674.safetensors", "step": 249674, "phase": "lookup_arm", "seed_offset": 7919, "tokens_per_step": 262144, "live_tokens_per_step": 262144, "weights": "the shipped bf16 checkpoint, upcast to the fp32 masters", "optimizers": "fresh (Muon momentum and Adam moments start at zero)", "guards_added": [], "utc": "2026-10-11T01:04:31Z"}, "probes": {"p0_grounding": {"acc": 0.36666666666666664, "families": {"count": 0.125, "permanence": 0.7142857142857143, "sequence": 0.125, "attribute": 0.5714285714285714}, "n": 30}, "p1_self_other": {"acc": 0.43333333333333335, "families": {"belief": 0.375, "reality": 0.6666666666666666, "perspective": 0.4, "reference": 0.5714285714285714, "second_order": 0.0}, "n": 30}, "p2_difference": {"acc": 0.5, "families": {"category": 0.8333333333333334, "antonym": 0.5, "odd_one_out": 0.5, "property": 0.5, "analogy": 0.16666666666666666}, "n": 30}, "p3_logic": {"acc": 0.2, "families": {"d1": 0.5, "d2": 0.0, "d3": 0.5, "d4": 0.25, "d5": 0.0, "d6": 0.0, "negation": 0.0, "contrapositive": 0.25}, "n": 30}, "p4_arithmetic": {"acc": 0.26666666666666666, "families": {"add2": 0.0, "add3": 0.0, "sub2": 0.5, "sub3": 0.4, "mul1": 0.25, "mul2": 0.3333333333333333, "word": 0.4}, "n": 30}, "p5_causal": {"acc": 0.6333333333333333, "families": {"effect": 0.7142857142857143, "cause": 0.42857142857142855, "counterfactual": 0.5, "prevention": 0.8, "chain": 0.8}, "n": 30}, "p6_learning": {"acc": 0.36666666666666664, "families": {"error_detect": 0.25, "correction": 0.25, "strategy": 0.7142857142857143, "check": 0.2857142857142857}, "n": 30}, "p7_composite": {"acc": 0.36666666666666664, "families": {"logic_arith": 0.25, "causal_category": 0.5, "belief_causal": 0.14285714285714285, "three_step": 0.5714285714285714}, "n": 30}, "p8_articulation": {"acc": 0.43333333333333335, "families": {"definition": 0.375, "register": 0.2857142857142857, "grammar": 0.875, "paraphrase": 0.14285714285714285}, "n": 30}}, "census_flags": {"hidden_erank_floor": true, "anchor_merge": false, "dispatch_collapse": false, "den_floor": false, "head_buried": false, "codebook_erank_floor": false}, "ledger": {"bpb_full": 0.9806994217374075, "bpb_bank_off": 5.481153468300945, "bpb_head_aleph_off": 2.863032638135401, "bpb_hub_off": 7.622422258989787, "toggle_bank_off": 4.500454046563537, "toggle_head_aleph_off": 1.8823332163979933, "toggle_hub_off": 6.64172283725238}, "special": {"doc_count": 70, "doc_bpb": 3.2413332184124752, "reset_bpb": 1.9765383681268875}}