© 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 EvalsExportsAPI 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. 6742
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ins" in various forms

    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
    616
    +0.14
    0.6%
    195
    +0.10
    0.5%
    735
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    616
    +0.14
    0.02
    195
    +0.10
    0.01
    744
    +0.09
    0.01
    Negative Logits
     XD
    -0.63
    ries
    -0.63
     Scotia
    -0.63
    llan
    -0.62
     Stafford
    -0.62
    ãĥ£
    -0.59
     Accountability
    -0.59
     Archdemon
    -0.58
     Legal
    -0.57
     Loving
    -0.57
    POSITIVE LOGITS
    urances
    1.18
    criptions
    1.10
    pite
    1.09
    ides
    1.07
    istent
    1.07
    ensible
    1.05
    uffer
    1.00
    incerity
    0.98
    uff
    0.95
    ensibly
    0.94
    Activations Density 0.031%

    No Known Activations