Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 19578
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     challeng
    -0.78
     misunder
    -0.75
     distingu
    -0.75
     irresist
    -0.75
     facult
    -0.74
     hemor
    -0.74
     trapping
    -0.72
     sacrific
    -0.72
     perpend
    -0.72
     exting
    -0.71
    POSITIVE LOGITS
    elta
    0.87
    dayName
    0.71
    aturday
    0.70
    essel
    0.69
    pod
    0.68
    pal
    0.67
    metadata
    0.66
    chan
    0.66
    clair
    0.66
    aples
    0.66
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.