Evaluation Architecture

A strict, three-layer evaluation framework with immutable frozen suites to ensure target capabilities are acquired without catastrophic forgetting of the baseline model's strengths.

Layer 1: Canonical Baseline Benchmark

Frozen

Immutable research anchor. Provides the primary comparison point against unmodified google/gemma-4-E4B-it.

Version
1.0.0
Test Cases
50

Layer 2: Capability Preservation Suite

Frozen

Monitors regression across baseline Gemma 4 strengths like teacher assistance, safety, and complex scientific explanation.

Version
1.0.0
Test Cases
30

Layer 3: Held-Out Behavioral Suite

Frozen

Measures whether fine-tuning acquired the 4 target behaviors: Socratic tutoring, concise calculation, clean translation, and Indic pedagogical tone.

Version
1.0.0
Test Cases
30

Strict Isolation Protocol

Under no circumstances may any case from these suites be included in, paraphrased for, or referenced during SFT or DPO dataset generation. Every proposed training batch must pass scripts/check_leakage.py with zero exact matches, zero parameter collisions, and < 35% n-gram overlap before inclusion.