© 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. 8364
    Prev
    Next
    INDEX
    Explanations

    locations or references to physical or metaphorical places

    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
    247
    +0.10
    0.5%
    82
    +0.09
    0.4%
    116
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    377
    +0.10
    0.06
    651
    +0.09
    0.05
    51
    +0.09
    0.04
    Negative Logits
    enced
    -0.67
    vez
    -0.65
    itate
    -0.62
    uously
    -0.62
    utive
    -0.61
     tongues
    -0.61
    cube
    -0.60
    Continue
    -0.60
    arget
    -0.59
    ertain
    -0.59
    POSITIVE LOGITS
    etsk
    0.76
    »Ĵ
    0.72
    ðŁij
    0.72
    ©¶æ¥µ
    0.71
     disadvant
    0.69
     srf
    0.69
    etheless
    0.67
    ntil
    0.67
     veter
    0.66
     spons
    0.66
    Activations Density 0.385%

    No Known Activations