© 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. 7230
    Prev
    Next
    INDEX
    Explanations

    terms associated with adherence and relational concepts

    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
    373
    +0.10
    0.4%
    559
    +0.10
    0.4%
    118
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.10
    0.04
    756
    +0.10
    -0.02
    575
    +0.09
    0.10
    Negative Logits
    dar
    -0.56
    _>
    -0.51
    kan
    -0.49
     lett
    -0.46
    INC
    -0.45
    fre
    -0.44
    owler
    -0.43
     Jou
    -0.42
    oufl
    -0.42
     grou
    -0.42
    POSITIVE LOGITS
     thereto
    1.44
     to
    1.31
    to
    1.10
    To
    1.04
     unto
    1.02
     To
    0.93
     TO
    0.91
    TO
    0.83
     toward
    0.69
     ta
    0.68
    Activations Density 2.044%

    No Known Activations