© 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. 4-RES-JB
    11. 4889
    Prev
    Next
    INDEX
    Explanations

    names of specific people, such as actors like Chris Pratt and directors like Steven Spielberg

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals, particularly associated with the film industry or notable events

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

    No Comments

    Negative Logits
     shelves
    -0.73
     gripping
    -0.69
     detainee
    -0.66
     meanings
    -0.64
     caut
    -0.63
     simultane
    -0.63
     vulner
    -0.62
     illnesses
    -0.62
     ailments
    -0.61
     basin
    -0.61
    POSITIVE LOGITS
     Pratt
    1.19
    issance
    1.02
    etry
    0.87
    son
    0.85
     Parsons
    0.84
    shaw
    0.83
    ulse
    0.78
    shire
    0.77
    acha
    0.76
    igham
    0.76
    Activations Density 0.019%

    No Known Activations