© 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. 0-RES-JB
    11. 9737
    Prev
    Next
    INDEX
    Explanations

    values that are important or enduring

    oai_token-act-pair · gpt-3.5-turbo

    references to moral or ethical principles

    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
    head
    -0.78
    chard
    -0.78
    horn
    -0.75
    aban
    -0.71
    linger
    -0.69
    jam
    -0.69
    lio
    -0.67
    cloth
    -0.67
    nec
    -0.66
    HEAD
    -0.66
    POSITIVE LOGITS
     values
    3.83
    values
    2.93
     Values
    2.92
    Values
    2.40
     value
    2.21
    value
    1.79
     Value
    1.75
     ideals
    1.73
    Value
    1.64
     coefficients
    1.42
    Activations Density 0.020%

    No Known Activations