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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 10-RES-JB
    6. 3624
    Prev
    Next
    INDEX
    Explanations

    terms related to subliminal messaging

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sub" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Muse
    -0.70
     Qiao
    -0.70
     goodbye
    -0.68
     Gates
    -0.66
     Bonds
    -0.65
     Cardinal
    -0.64
     Forever
    -0.63
     Patron
    -0.63
     Bills
    -0.63
     Dragonbound
    -0.62
    POSITIVE LOGITS
    missive
    1.45
    mits
    1.32
    verting
    1.32
    mitter
    1.31
    culture
    1.30
    species
    1.24
    sequent
    1.22
    verted
    1.22
    continental
    1.21
    sequently
    1.21
    Activations Density 0.016%

    No Known Activations