top | item 44858175 (no title) poorman | 6 months ago As we saw with GPT-5 the RL technique of training doesn't scale forever discuss order hn newest energy123|6 months ago Unless GPT-5 is 30% cheaper to run than o3. Then it's scaling brilliantly given the small gap between release dates. People are really drawing too many conclusions from too little information. oezi|6 months ago I meant scaling the base training before RL.
energy123|6 months ago Unless GPT-5 is 30% cheaper to run than o3. Then it's scaling brilliantly given the small gap between release dates. People are really drawing too many conclusions from too little information.
energy123|6 months ago
oezi|6 months ago