You are coding as a Guest. Sign in with your RoleNest account to permanently track your streak, earn XP, and climb the Campus Leaderboard!
Sign In with RoleNestProblem Set
🔥Stable Softmax & Temperature (LLM Sampling)Medium
DAILY PROBLEM OF THE DAY+50 XP • Daily Streak
Solve today's challenge or tackle one of the 3 Super Hard challenges for +150 XP.
MediumAI & Machine Learning•Acceptance: 58.1%
Stable Softmax & Temperature (LLM Sampling)
Targeted in FAANG & Tech OA:GoogleAmazonMicrosoftSwiggy
Real-World Engineering Context
Directly implements the final token sampling layer of ChatGPT and Llama-3, translating raw transformer unnormalized log-probabilities into categorical next-token predictions.
Given a 1D array of float `logits` and a positive `temperature`, compute the numerically stable Softmax probability distribution. Each element should be rounded to 4 decimal places:
$$P(i) = \frac{\exp((\text{logits}[i] - \max(\text{logits})) / T)}{\sum_j \exp((\text{logits}[j] - \max(\text{logits})) / T)}$$
Sample Test Cases
Input: [[10,10],1]
Expected: [0.5,0.5]
Input: [[1000,1000],1]
Expected: [0.5,0.5]
Constraints
- 1 <= logits.length <= 1000
- -1000.0 <= logits[i] <= 1000.0
- 0.1 <= temperature <= 5.0
Recruiter Fast-Track ReferralVerified Candidate
Direct pipeline to Google, Amazon, Microsoft, Swiggy, & Uber recruiters
Top DevScore profiles bypass resume screening filters. Every verified problem solve writes authentic proof-of-work to your profile and dispatches you directly into employer inboxes on RoleNest.
Explore 5,000+ Direct ATS Openings on RoleNest ↗✓ Direct Referral Active
Language:
Ready to test. Click Run Code or Submit Solution to run test cases in isolated browser sandbox.