© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 4970
    Prev
    Next
    INDEX
    Explanations

    terms related to unethical or illegal activities such as fraud, manipulation, corruption, and coercion

    oai_token-act-pair · gpt-3.5-turbo

    terms related to assessments, potential threats, and manipulative promises

    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
    undrum
    -0.67
    NER
    -0.66
    ottage
    -0.64
    ipeg
    -0.63
    issance
    -0.61
    eworks
    -0.60
    ipment
    -0.58
    yssey
    -0.58
     predicament
    -0.58
    ventures
    -0.58
    POSITIVE LOGITS
    Downloadha
    1.16
     thereof
    1.14
     resembling
    0.97
     of
    0.88
    ensitive
    0.76
     unfavorable
    0.76
    pread
    0.75
     conducive
    0.73
     hostile
    0.73
     deemed
    0.71
    Activations Density 0.481%

    No Known Activations