lomahony/pythia-1.4b-helpful-dpo
Pythia-1.4b DPO finetuned using original DPO code with the helpful subset of Anthropic-hh-rlhf dataset for 1 epoch.
Checkpoints are also uploaded.
Fully reproducible finetuning code is available on GitHub
See Pythia-1.4b for model details (paper).
See further details of these models in the paper Attributing Mode Collapse in the Fine-Tuning of Large Language Models.
You can cite these models if they are helpful as follows:
<pre> @inproceedings{o2024attributing, title={Attributing Mode Collapse in the Fine-Tuning of Large Language Models}, author={O’Mahony, Laura and Grinsztajn, Leo and Schoelkopf, Hailey and Biderman, Stella}, booktitle={ICLR 2024, Mathematical and Empirical Understanding of Foundation Models (ME-FoMo) workshop}, year={2024} } </pre>
hf (pretrained=lomahony/pythia-1.4b-helpful-dpo), genkwargs: (None), limit: None, numfewshot: 0, batchsize: 16 | Tasks |Version|Filter|n-shot| Metric | Value | |Stderr| |--------------|------:|------|-----:|---------------|------:|---|------| |arcchallenge | 1|none | 0|acc | 0.2816|± |0.0131| | | |none | 0|accnorm | 0.3123|± |0.0135| |arceasy | 1|none | 0|acc | 0.6229|± |0.0099| | | |none | 0|accnorm | 0.5459|± |0.0102| |boolq | 2|none | 0|acc | 0.6229|± |0.0085| |hellaswag | 1|none | 0|acc | 0.4191|± |0.0049| | | |none | 0|accnorm | 0.5383|± |0.0050| |lambadaopenai| 1|none | 0|perplexity | 6.4790|± |0.1947| | | |none | 0|acc | 0.5674|± |0.0069| |openbookqa | 1|none | 0|acc | 0.2280|± |0.0188| | | |none | 0|accnorm | 0.3360|± |0.0211| |piqa | 1|none | 0|acc | 0.7122|± |0.0106| | | |none | 0|accnorm | 0.7214|± |0.0105| |sciq | 1|none | 0|acc | 0.8480|± |0.0114| | | |none | 0|accnorm | 0.7840|± |0.0130| |wikitext | 2|none | 0|wordperplexity|16.4022|± |N/A | | | |none | 0|byteperplexity| 1.6873|± |N/A | | | |none | 0|bitsperbyte | 0.7547|± |N/A | |winogrande | 1|none | 0|acc | 0.5959|± |0.0138|
hf (pretrained=lomahony/pythia-1.4b-helpful-dpo), genkwargs: (None), limit: None, numfewshot: 5, batchsize: 16 | Tasks |Version|Filter|n-shot| Metric | Value | |Stderr| |--------------|------:|------|-----:|---------------|------:|---|------| |arcchallenge | 1|none | 5|acc | 0.3089|± |0.0135| | | |none | 5|accnorm | 0.3353|± |0.0138| |arceasy | 1|none | 5|acc | 0.6423|± |0.0098| | | |none | 5|accnorm | 0.6334|± |0.0099| |boolq | 2|none | 5|acc | 0.6291|± |0.0084| |hellaswag | 1|none | 5|acc | 0.4124|± |0.0049| | | |none | 5|accnorm | 0.5347|± |0.0050| |lambadaopenai| 1|none | 5|perplexity | 9.7688|± |0.3083| | | |none | 5|acc | 0.4904|± |0.0070| |openbookqa | 1|none | 5|acc | 0.2260|± |0.0187| | | |none | 5|accnorm | 0.3240|± |0.0210| |piqa | 1|none | 5|acc | 0.7095|± |0.0106| | | |none | 5|accnorm | 0.7165|± |0.0105| |sciq | 1|none | 5|acc | 0.9140|± |0.0089| | | |none | 5|accnorm | 0.9050|± |0.0093| |wikitext | 2|none | 5|wordperplexity|16.4022|± |N/A | | | |none | 5|byteperplexity| 1.6873|± |N/A | | | |none | 5|bitsperbyte | 0.7547|± |N/A | |winogrande | 1|none | 5|acc | 0.5612|± |0.0139|