Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 8882
    Prev
    Next
    INDEX
    Explanations

    phrases related to referrals or reference requests

    oai_token-act-pair · gpt-3.5-turbo

    terms related to referrals in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     emperor
    -0.78
    hess
    -0.74
     Empress
    -0.72
     Sophie
    -0.70
     Aust
    -0.69
     havoc
    -0.68
     Dav
    -0.67
    shape
    -0.66
     Monstrous
    -0.65
    ATURES
    -0.64
    POSITIVE LOGITS
     referral
    3.39
     referrals
    3.09
     Refer
    2.56
    Refer
    2.02
     vouchers
    1.42
     voucher
    1.32
     reimbursement
    1.21
     redirect
    1.21
    erred
    1.21
     Referred
    1.18
    Activations Density 0.045%

    No Known Activations