© 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. 26993
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "hit" in various 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
    243
    +0.11
    0.5%
    366
    +0.10
    0.5%
    125
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    526
    +0.11
    -0.01
    125
    +0.10
    0.02
    366
    +0.10
    0.02
    Negative Logits
    gio
    -0.67
    urdue
    -0.64
    construct
    -0.64
     Conversation
    -0.63
    iterator
    -0.60
    lege
    -0.60
     UID
    -0.59
    umn
    -0.59
    fed
    -0.58
    aughter
    -0.58
    POSITIVE LOGITS
     stride
    1.26
     snag
    1.05
     button
    0.94
     itch
    0.92
     milestones
    0.91
     nail
    0.89
     hardest
    0.88
     bumps
    0.88
     targets
    0.86
     mark
    0.86
    Activations Density 0.603%

    No Known Activations