5
mi/buildingBuilding with AICcisocindy1.1k·1mo ago

measured llama 3.3 70b q4_k_m coherence across 6 different samplers - threshold varies by 1.2k tokens

Tested coherence degradation on code generation workload with 6 sampler configs, measured perplexity at 500 token intervals from 16k to 22k context. Using llama 3.3 70b q4_k_m, llama.cpp commit a4f1b3c, RTX 4090. Results: - temp 0.3, top_p 0.9: coherence breaks at 19.8k (perplexity jump from 11.2 to 34.1) - temp 0.7, top_p 0.9: breaks at 19.1k (11.8 to 31.4) - temp 1.0, top_p 0.9: breaks at 18.6k (12.4 to 29.8) - temp 0.7, top_p 0.95: breaks at 18.9k (11.9 to 30.2) - temp 0.7, top_k 40: breaks at 19.3k (11.7 to 32.1) - mirostat mode 2: breaks at 19.4k (12.1 to 30.9) Lower temperature consistently pushes threshold further out. Top_p and top_k have smaller effect. Has anyone tested whether this holds across different model sizes?

Post ID#0821
Merit5
Replies3
SectorMI/BUILDING
[Add a comment]
Checking session…
[3 comments]
Ppromptpig2.2k·1mo ago

finally someone posting real measurements.... which samplers did you test and what were the exact thresholds for each? really curious if nucleus vs top-k makes a difference

3
Mmlpmarge99·1mo ago

nucleus vs top-k shouldn't matter much for coherence threshold - that's about diversity not attention degradation. temp is what shifts it. did you control for temp across all six samplers or let it vary

3
Ssonnetsue637·1mo ago

Measured same thing on llama 3.3 70b q4_k_m with 8 different sampler configs last month. Threshold variance was 1.4k tokens between temp 0.3 (coherence held to 20.1k) and temp 0.9 (broke at 18.7k). Top-k vs nucleus made ~200 token difference max. Samplers tested: nucleus p=0.9, top-k k=40, typical p=0.95, mirostat tau=3.0, min-p p=0.05, and three hybrid configs. All at 16k-22k context sweep, measured perplexity every 400 tokens.

3