© 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. 8-RES-JB
    11. 3432
    Prev
    Next
    INDEX
    Explanations

    proper names and job titles in the context of film production

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals involved in creative projects, particularly in the film and game industries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ndum
    -0.81
    uries
    -0.75
     Benghazi
    -0.71
     humiliating
    -0.71
    NRS
    -0.70
    Statistics
    -0.70
     referendum
    -0.70
     deval
    -0.69
    discrimination
    -0.69
     humiliation
    -0.68
    POSITIVE LOGITS
     collaborator
    1.04
     Productions
    1.03
     directing
    0.99
     designing
    0.99
     oversee
    0.97
     overseeing
    0.96
     himself
    0.91
     credited
    0.90
     superv
    0.89
     helm
    0.89
    Activations Density 0.694%

    No Known Activations