© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 7585
    Prev
    Next
    INDEX
    Explanations

    terms related to moderate-sized sports or activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    314
    +0.12
    0.5%
    163
    +0.12
    0.5%
    457
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    0.01
    167
    +0.12
    0.02
    288
    +0.11
    0.16
    Negative Logits
     dots
    -0.71
     needles
    -0.67
     Pluto
    -0.64
    zzle
    -0.62
     Rasmussen
    -0.62
    ĸļ
    -0.62
     ['
    -0.61
     torped
    -0.61
     Cabrera
    -0.60
    luaj
    -0.60
    POSITIVE LOGITS
    hooting
    0.79
    atform
    0.78
    bernatorial
    0.77
    inese
    0.74
    ophile
    0.73
    ean
    0.72
     therap
    0.72
    iciary
    0.71
    Tech
    0.71
    ollywood
    0.68
    Activations Density 2.754%

    No Known Activations