© 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. 10-RES_POST_32K-OAI
    11. 32159
    Prev
    Next
    INDEX
    Explanations

    references to institutions or organizations

    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
    373
    +0.23
    1.0%
    412
    +0.13
    0.6%
    39
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.23
    0.13
    495
    +0.13
    0.08
    651
    +0.10
    0.10
    Negative Logits
    haar
    -0.76
    ystem
    -0.71
    Ô
    -0.69
    blance
    -0.68
     Assignment
    -0.67
    poon
    -0.66
    vation
    -0.65
    ynthesis
    -0.64
    lap
    -0.63
    mith
    -0.63
    POSITIVE LOGITS
     itself
    1.18
    osphere
    0.95
    iverse
    0.81
     proper
    0.68
     formerly
    0.66
     perimeter
    0.66
    's
    0.65
     hierarchy
    0.65
     treasury
    0.63
     themselves
    0.61
    Activations Density 1.203%

    No Known Activations