© 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. 3-RES-JB
    11. 3672
    Prev
    Next
    INDEX
    Explanations

    text related to inviting or joining events or groups

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to invitations and gatherings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    phia
    -0.64
    inson
    -0.63
    gran
    -0.62
    complex
    -0.61
     impacts
    -0.61
     fueled
    -0.59
     Phen
    -0.58
     triggered
    -0.58
     disproportion
    -0.58
     deficient
    -0.58
    POSITIVE LOGITS
     fray
    1.13
    Ô
    0.82
    orate
    0.79
    ername
    0.79
    EStreamFrame
    0.77
     Downing
    0.77
    oration
    0.76
    addle
    0.71
     bandwagon
    0.71
    ATTLE
    0.69
    Activations Density 0.261%

    No Known Activations