Add code long-context attention profiling grid
This commit is contained in:
@@ -33,7 +33,9 @@ max_model_len_recommendation != null
|
||||
selected.selected_window_stats.max_model_len_coverage[推荐值].coverage = 1.0
|
||||
```
|
||||
|
||||
旧记录预计 source block size 为 512、推荐 max model len 为 131072,但禁止把这两个值写死为实验事实。
|
||||
全量审计已确认 source block size=512;development source window 若 100%
|
||||
覆盖需要 262144,但正式 server cap 以 session-sampled paired cell 的实际
|
||||
`ISL+OSL max` 向上对齐,不能把 full-window 262144 无条件套到低 rho cell。
|
||||
审计会单独记录并排除 `input_length<=0` 或 `output_length<=0` 的 source
|
||||
行;这些行只有在 raw trace 同样显示 zero usage/empty response 时才按
|
||||
“未发生模型执行”处理,不能无记录过滤。
|
||||
@@ -89,14 +91,14 @@ python3 runs/frontier-s3-real-v0/remap_hash_blocks.py \
|
||||
现有 real runner 新增了三个显式环境变量,chat 默认行为不变:
|
||||
|
||||
```bash
|
||||
MAX_MODEL_LEN=131072 \
|
||||
MAX_MODEL_LEN=ACTUAL_CELL_MAX_ROUNDED_UP \
|
||||
TRACE_INPUT_ROOT=/absolute/path/to/materialized/code-cell \
|
||||
ALLOW_SYNTHETIC_PROMPTS=true \
|
||||
OUTPUT_ROOT=/absolute/path/to/new/output \
|
||||
bash runs/frontier-s3-real-v0/run_full_real.sh RHO_LABEL tp4_mns16 1 PORT
|
||||
```
|
||||
|
||||
- `MAX_MODEL_LEN` 必须等于 manifest 推荐值;
|
||||
- `MAX_MODEL_LEN` 必须覆盖 manifest 中该 paired cell 的实际最大请求;
|
||||
- `TRACE_INPUT_ROOT` 内必须有 `real_requests.jsonl` 和 `manifest.json`;
|
||||
- synthetic prompt 默认拒绝,只有在 experiment card 明确降级 claim 后才设为 `true`;
|
||||
- runner 会在启动前扫描 paired requests,若任何 `ISL+OSL` 超 cap 立即失败。
|
||||
|
||||
Reference in New Issue
Block a user