13 Commits

Author	SHA1	Message	Date
Richie Cahill	26105b7daa	updated BenchmarkConfig to have from_toml	2026-04-10 21:55:18 -04:00
Richie Cahill	0d81f2d17b	setup FinetuneConfig	2026-04-10 21:40:17 -04:00
Richie Cahill	1409e9c63e	deleted train.sh	2026-04-10 20:58:26 -04:00
Richie Cahill	259e952afc	added containers dir	2026-04-10 20:48:24 -04:00
Richie Cahill	4a10a80ba0	conveted to summarization_prompts	2026-04-10 18:57:21 -04:00
Richie Cahill	03208a1ab2	moved renamed container.py to vllm_container.py	2026-04-10 13:16:18 -04:00
Richie Cahill	721526022b	created working finetuing pipeline	2026-04-10 12:56:57 -04:00
Richie Cahill	88dae310b6	added summarization_prompts.py to sore the prompts	2026-04-10 12:40:36 -04:00
Richie Cahill	24f0e8693a	added tools dir for on off scripts i used	2026-04-10 12:37:14 -04:00
Richie Cahill	ced78fe516	added batch_bill_summarizer.py ... batch bill summarizer sends a batch api call to gpt	2026-04-10 12:36:39 -04:00
Richie Cahill	251da6c14a	added bill_token_compression.py ... tested on sample size of 100 bills matching the distribution of our data Compression saves ~11.5% on prompt tokens; completion/reasoning are roughly equal across the two sets. prompt completion reasoning total compressed 349,460 157,110 112,128 506,570 uncompressed 394,948 154,710 110,080 549,658 delta −45,488 +2,400 +2,048 −43,088	2026-04-09 18:41:13 -04:00
Richie Cahill	d17c883476	created main prompt bench	2026-04-08 09:08:25 -04:00
Richie Cahill	9c8013d69d	creating prompt_bench downloader	2026-04-07 19:15:43 -04:00