© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 29846
    Prev
    Next
    INDEX
    Explanations

    technical components and their connections in hardware contexts

    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
    535
    +0.11
    0.5%
    500
    +0.10
    0.5%
    200
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.06
    326
    +0.10
    0.05
    41
    +0.09
    0.04
    Negative Logits
    ichick
    -0.81
    tsky
    -0.81
    castle
    -0.73
    Imm
    -0.72
    fare
    -0.70
    opa
    -0.69
     sovere
    -0.69
    spell
    -0.68
    itia
    -0.67
     Archdemon
    -0.64
    POSITIVE LOGITS
     cables
    1.05
     cable
    0.94
     wires
    0.90
     sockets
    0.85
     adapter
    0.84
     wiring
    0.84
     adapters
    0.83
     socket
    0.83
     piping
    0.82
     plugs
    0.77
    Activations Density 0.332%

    No Known Activations