You can deploy this model and start earning money today!

jchang153/
qwen25-7b-humor-dpo-lls-random-80

Name: qwen25-7b-humor-dpo-lls-random-80
Availability: InStock

Qwen/Qwen2.5-7B-Instruct

adapter

Views

No views yet

Qwen2.5-7B humor DPO LoRA: random_80

This is the random_80 arm of the targeted-LLS humor DPO experiment. It is a private, training-only research artifact. Behavioral evaluation is intentionally deferred, so this model card makes no behavioral mitigation claim. The source preference data remains subject to its non-commercial/research terms.

Provenance

Base model revision: a09a35458c702b33eeacc393d103063234e8bc28
OpenCharacterTraining revision: d1da9f03628cb4c5482ba2e494a7cba33bcd5818
OpenRLHF revision: eaf40e10e0471a9e50d33697bcef15f7b0a32b05
Dataset manifest SHA-256: e59433a173107f7cfc901234a64794c2dd0cb447c00a13252facfece2a846d8b
Arm dataset SHA-256: dc5d09998ec6a3147783c566e2c3002c13e78eb1a26671b35fb4b3bf4dd98099
LoRA rank: 64
LoRA alpha: 128

Local adapter hashes

adapter_config.json: bc43809a08b38b6c08116b7e21336e44968d892f960a760b345cb8ab7f63d295
adapter_model.safetensors: a1f25e1e9c48d384ef92ed9729226c863b4b78256b37f22862a0ccd38b8fc44d

jchang153/qwen25-7b-humor-dpo-lls-random-80

Qwen2.5-7B humor DPO LoRA: random_80

Provenance

Local adapter hashes

jchang153/
qwen25-7b-humor-dpo-lls-random-80