© 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. 10-RES_POST_32K-OAI
    11. 29246
    Prev
    Next
    INDEX
    Explanations

    instances of the word "hard" associated with effort and perseverance

    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
    459
    +0.11
    0.5%
    247
    +0.11
    0.5%
    761
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    247
    +0.11
    0.03
    51
    +0.11
    0.03
    271
    +0.11
    0.03
    Negative Logits
    eele
    -0.68
    DragonMagazine
    -0.65
     rul
    -0.59
    URR
    -0.58
    Closure
    -0.57
    cffffcc
    -0.57
    ij士
    -0.56
    CVE
    -0.56
     transsexual
    -0.56
    obyl
    -0.55
    POSITIVE LOGITS
     enough
    1.07
    enough
    0.91
    terday
    0.86
     throughout
    0.81
    night
    0.76
    eday
    0.76
    entimes
    0.74
     every
    0.72
    mouth
    0.72
    nat
    0.72
    Activations Density 0.235%

    No Known Activations