Skip to page content
SAVRN
Search Plan a project

Open-weight model

monitor_cot_for_hack-reward_description-length-explicit_intent-gpt-oss-20b

by John's Unnamed Lab hewittlab/monitor_cot_for_hack-reward_description-length-explicit_intent-gpt-oss-20b

monitor_cot_for_hack-reward_description-length-explicit_intent-gpt-oss-20b is an open-weight model from John's Unnamed Lab. Its published files total 69.5 GB.

LoRA checkpoints from Reward Misspecification Degrades Chain-of-Thought Monitorability (code).

Parameters—
Context—
Weights69.5 GB
License—
AccessOpen weights
Monthly Downloads—

Model Card

LoRA checkpoints from Reward Misspecification Degrades Chain-of-Thought Monitorability (code). - controlmodel: trained on the true reward: 1 if the final output has at most 909 words split by whitespace, else 0 - misalignedmodel: trained on the misspecified reward: 1 if the final output has more than 909 words split by whitespace, else 0 The adapters are in Tinker's format. Merge one into the base model with tinker-cookbook (see Tinker's Export to HF guide), then load it The base model is downloaded without its original/ and metal/ folders, which the merge does not use.

Excerpt from the card by John's Unnamed Lab.

Identity and Version

Repository
hewittlab/monitor_cot_for_hack-reward_description-length-explicit_intent-gpt-oss-20b
Publisher
John's Unnamed Lab
Task
Not stated by the source
Modality
Other
Library
Not stated by the source
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
cbb0f926868e9bba006361629beeb74afb082430
First published
2026-10-16
Last updated
2026-10-17

Files and Weights

158 files, 69.5 GB in total. The weights are 72 files totalling 69.5 GB in safetensors.

Weights72 files · 69.5 GB
Configuration78 files · 79.3 KB
Documentation1 file · 2.9 KB
Other6 files · 1.1 KB
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
seed_0/control_model/step_000010/adapter_model.safetensorsWeights965.7 MB febb4120f256
seed_0/control_model/step_000020/adapter_model.safetensorsWeights965.7 MB 2fa1fdd1add7
seed_0/control_model/step_000030/adapter_model.safetensorsWeights965.7 MB 6e6c1a974899
seed_0/control_model/step_000040/adapter_model.safetensorsWeights965.7 MB b7dcba765fe5
seed_0/control_model/step_000050/adapter_model.safetensorsWeights965.7 MB 663d9801c0cb
seed_0/control_model/step_000060/adapter_model.safetensorsWeights965.7 MB 9ce61963aa08
seed_0/control_model/step_000070/adapter_model.safetensorsWeights965.7 MB 8cd652ecf14a
seed_0/control_model/step_000080/adapter_model.safetensorsWeights965.7 MB 6206b152d171
seed_0/control_model/step_000090/adapter_model.safetensorsWeights965.7 MB 5337436d61ef
seed_0/control_model/step_000100/adapter_model.safetensorsWeights965.7 MB 4fa5631f4a53
seed_0/control_model/step_000110/adapter_model.safetensorsWeights965.7 MB 4fa5631f4a53
seed_0/control_model/step_000120/adapter_model.safetensorsWeights965.7 MB 2119b48f052e
seed_0/misaligned_model/step_000010/adapter_model.safetensorsWeights965.7 MB a4dda62c54cd
seed_0/misaligned_model/step_000020/adapter_model.safetensorsWeights965.7 MB a19997700129
seed_0/misaligned_model/step_000030/adapter_model.safetensorsWeights965.7 MB 6530b13ce117
seed_0/misaligned_model/step_000040/adapter_model.safetensorsWeights965.7 MB 8358cebd94a2
seed_0/misaligned_model/step_000050/adapter_model.safetensorsWeights965.7 MB ffcee52bd448
seed_0/misaligned_model/step_000060/adapter_model.safetensorsWeights965.7 MB ab9bc6111a78
seed_0/misaligned_model/step_000070/adapter_model.safetensorsWeights965.7 MB 28344d9dbfcd
seed_0/misaligned_model/step_000080/adapter_model.safetensorsWeights965.7 MB 62f5ee7f8f3c
seed_0/misaligned_model/step_000090/adapter_model.safetensorsWeights965.7 MB 5ede40e64d9c
seed_0/misaligned_model/step_000100/adapter_model.safetensorsWeights965.7 MB c6834fed7aa0
seed_0/misaligned_model/step_000110/adapter_model.safetensorsWeights965.7 MB c6834fed7aa0
seed_0/misaligned_model/step_000120/adapter_model.safetensorsWeights965.7 MB c6834fed7aa0
seed_1/control_model/step_000010/adapter_model.safetensorsWeights965.7 MB 40455acefc1e
seed_1/control_model/step_000020/adapter_model.safetensorsWeights965.7 MB bf1446f57ead
seed_1/control_model/step_000030/adapter_model.safetensorsWeights965.7 MB 2fcec8ada5ae
seed_1/control_model/step_000040/adapter_model.safetensorsWeights965.7 MB 1d7b972a6afb
seed_1/control_model/step_000050/adapter_model.safetensorsWeights965.7 MB c2079a32e7cf
seed_1/control_model/step_000060/adapter_model.safetensorsWeights965.7 MB d8ce33ca5524
seed_1/control_model/step_000070/adapter_model.safetensorsWeights965.7 MB d5702b672b1c
seed_1/control_model/step_000080/adapter_model.safetensorsWeights965.7 MB 3596e77f9894
seed_1/control_model/step_000090/adapter_model.safetensorsWeights965.7 MB 31d2c0bff98e
seed_1/control_model/step_000100/adapter_model.safetensorsWeights965.7 MB 05b860253ee0
seed_1/control_model/step_000110/adapter_model.safetensorsWeights965.7 MB b6828fb24756
seed_1/control_model/step_000120/adapter_model.safetensorsWeights965.7 MB eb5350c69175
seed_1/misaligned_model/step_000010/adapter_model.safetensorsWeights965.7 MB 2f23904d3c1d
seed_1/misaligned_model/step_000020/adapter_model.safetensorsWeights965.7 MB 2796173f719a
seed_1/misaligned_model/step_000030/adapter_model.safetensorsWeights965.7 MB 092f8371551f
seed_1/misaligned_model/step_000040/adapter_model.safetensorsWeights965.7 MB aa3f90cbea68
seed_1/misaligned_model/step_000050/adapter_model.safetensorsWeights965.7 MB 804455a120d9
seed_1/misaligned_model/step_000060/adapter_model.safetensorsWeights965.7 MB 844f9f681f9c
seed_1/misaligned_model/step_000070/adapter_model.safetensorsWeights965.7 MB 844f9f681f9c
seed_1/misaligned_model/step_000080/adapter_model.safetensorsWeights965.7 MB fe22caf331c0
seed_1/misaligned_model/step_000090/adapter_model.safetensorsWeights965.7 MB fe22caf331c0
seed_1/misaligned_model/step_000100/adapter_model.safetensorsWeights965.7 MB 702e7a4ed138
seed_1/misaligned_model/step_000110/adapter_model.safetensorsWeights965.7 MB 702e7a4ed138
seed_1/misaligned_model/step_000120/adapter_model.safetensorsWeights965.7 MB 702e7a4ed138
seed_2/control_model/step_000010/adapter_model.safetensorsWeights965.7 MB b0fb70f84ff6
seed_2/control_model/step_000020/adapter_model.safetensorsWeights965.7 MB c4276d845ca1
seed_2/control_model/step_000030/adapter_model.safetensorsWeights965.7 MB ea3929226f9c
seed_2/control_model/step_000040/adapter_model.safetensorsWeights965.7 MB 72bc0525de2c
seed_2/control_model/step_000050/adapter_model.safetensorsWeights965.7 MB 874f112d2552
seed_2/control_model/step_000060/adapter_model.safetensorsWeights965.7 MB cbaae926a537
seed_2/control_model/step_000070/adapter_model.safetensorsWeights965.7 MB 80d62059d02f
seed_2/control_model/step_000080/adapter_model.safetensorsWeights965.7 MB 79f84f8b4bfb
seed_2/control_model/step_000090/adapter_model.safetensorsWeights965.7 MB 0e433a3b4398
seed_2/control_model/step_000100/adapter_model.safetensorsWeights965.7 MB b31aa7013637
seed_2/control_model/step_000110/adapter_model.safetensorsWeights965.7 MB ebdb0cc2b2d3
seed_2/control_model/step_000120/adapter_model.safetensorsWeights965.7 MB ebdb0cc2b2d3
seed_2/misaligned_model/step_000010/adapter_model.safetensorsWeights965.7 MB faec5a0fdc50
seed_2/misaligned_model/step_000020/adapter_model.safetensorsWeights965.7 MB 25fb86e37c96
seed_2/misaligned_model/step_000030/adapter_model.safetensorsWeights965.7 MB 72e588f4a6b0
seed_2/misaligned_model/step_000040/adapter_model.safetensorsWeights965.7 MB 885461f31cf8
seed_2/misaligned_model/step_000050/adapter_model.safetensorsWeights965.7 MB 4b1a7fdcd67a
seed_2/misaligned_model/step_000060/adapter_model.safetensorsWeights965.7 MB b757a4fb19dc
seed_2/misaligned_model/step_000070/adapter_model.safetensorsWeights965.7 MB ca108d543652
seed_2/misaligned_model/step_000080/adapter_model.safetensorsWeights965.7 MB ec65cef1a606
seed_2/misaligned_model/step_000090/adapter_model.safetensorsWeights965.7 MB 4a0ea052c192
seed_2/misaligned_model/step_000100/adapter_model.safetensorsWeights965.7 MB 4a0ea052c192
seed_2/misaligned_model/step_000110/adapter_model.safetensorsWeights965.7 MB d1212913cbcc
seed_2/misaligned_model/step_000120/adapter_model.safetensorsWeights965.7 MB 260190a34648
seed_0/control_model/config.yamlConfiguration2.0 KB —
seed_0/control_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_0/control_model/step_000120/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/config.yamlConfiguration2.0 KB —
seed_0/misaligned_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_0/misaligned_model/step_000120/adapter_config.jsonConfiguration934 B —
seed_1/control_model/config.yamlConfiguration2.0 KB —
seed_1/control_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_1/control_model/step_000120/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/config.yamlConfiguration2.0 KB —
seed_1/misaligned_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_1/misaligned_model/step_000120/adapter_config.jsonConfiguration934 B —
seed_2/control_model/config.yamlConfiguration2.0 KB —
seed_2/control_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_2/control_model/step_000120/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/config.yamlConfiguration2.0 KB —
seed_2/misaligned_model/step_000010/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000020/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000030/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000040/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000050/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000060/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000070/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000080/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000090/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000100/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000110/adapter_config.jsonConfiguration934 B —
seed_2/misaligned_model/step_000120/adapter_config.jsonConfiguration934 B —
README.mdDocumentation2.9 KB —
seed_0/control_model/reward_description.txtOther123 B —
seed_0/misaligned_model/reward_description.txtOther193 B —
seed_1/control_model/reward_description.txtOther153 B —
seed_1/misaligned_model/reward_description.txtOther265 B —
seed_2/control_model/reward_description.txtOther131 B —
seed_2/misaligned_model/reward_description.txtOther189 B —
.gitattributesRepository1.5 KB —

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
69.5 GB
Download from John's Unnamed Lab

Released by John's Unnamed Lab through its official repository on Hugging Face.

Built From

Memory Requirements

PrecisionWeights in memory
As published69.5 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.