© 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. 27454
    Prev
    Next
    INDEX
    Explanations

    occurrences of the substring "sh" in words

    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
    22
    +0.09
    0.4%
    378
    +0.08
    0.4%
    459
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.09
    0.04
    232
    +0.08
    0.04
    314
    +0.08
    -0.00
    Negative Logits
    ulia
    -0.83
    issan
    -0.68
     libel
    -0.63
     occurrences
    -0.63
     embold
    -0.61
    idian
    -0.60
    anwhile
    -0.60
     paraph
    -0.60
     Rebell
    -0.59
     sources
    -0.59
    POSITIVE LOGITS
    oe
    1.12
    irted
    0.99
    oes
    0.98
    aker
    0.91
    orts
    0.82
    rimp
    0.82
    iners
    0.81
    aft
    0.81
    IRT
    0.80
    ave
    0.79
    Activations Density 0.382%

    No Known Activations