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

    Neuronpedia

    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. 13784
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Horn" with varying degrees of activation

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Horn" and its variants, often 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
    ãĥ¥
    -0.69
     MIA
    -0.67
    amin
    -0.65
    ENCY
    -0.64
    erate
    -0.64
     Pepsi
    -0.63
    ufact
    -0.63
    ensional
    -0.61
    BILITIES
    -0.61
    perature
    -0.61
    POSITIVE LOGITS
    et
    0.99
    obyl
    0.96
    worms
    0.90
    ado
    0.89
    sey
    0.87
    worm
    0.87
    iasis
    0.85
    entimes
    0.83
    beam
    0.82
    stein
    0.82
    Activations Density 0.044%

    No Known Activations