© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 31382
    Prev
    Next
    INDEX
    Explanations

    conditional statements or clauses that begin with "if."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    757
    +0.10
    0.4%
    108
    +0.09
    0.4%
    163
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.10
    0.08
    59
    +0.09
    0.06
    572
    +0.09
    0.08
    Negative Logits
     regress
    -0.77
    improve
    -0.65
     regression
    -0.65
     disadvant
    -0.64
     altogether
    -0.63
     catast
    -0.63
     Apart
    -0.61
     releg
    -0.61
     relegation
    -0.61
     egreg
    -0.61
    POSITIVE LOGITS
     wearer
    0.79
     purchased
    0.79
     convicted
    0.77
     registered
    0.73
     booked
    0.73
     buyer
    0.73
    ername
    0.72
     completed
    0.71
    pleted
    0.69
     contestants
    0.69
    Activations Density 0.923%

    No Known Activations