jkminder's picture
Upload Meta-Llama-3.1-8B-L16-mu2.0e-02-lr1e-04-local-shuffling-CCLoss dictionary model
2360181 verified
{
"trainer": {
"dict_class": "CrossCoder",
"trainer_class": "CrossCoderTrainer",
"activation_dim": 4096,
"dict_size": 131072,
"lr": 0.0001,
"l1_penalty": 0.02,
"warmup_steps": 1000,
"resample_steps": null,
"device": "cuda",
"layer": 16,
"lm_name": "meta-llama/Meta-Llama-3.1-8B-Instruct-meta-llama/Meta-Llama-3.1-8B",
"wandb_name": "Meta-Llama-3.1-8B-L16-mu2.0e-02-lr1e-04-local-shuffling-CCLoss",
"submodule_name": null,
"use_mse_loss": false,
"code_normalization": "CROSSCODER",
"code_normalization_alpha_sae": 1.0,
"code_normalization_alpha_cc": 0.1
}
}