© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 4713
    Prev
    Next
    INDEX
    Explanations

    words associated with fantastical or mythical creatures

    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
    274
    +0.15
    0.7%
    496
    +0.12
    0.6%
    119
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.02
    284
    +0.12
    0.01
    151
    +0.11
    0.01
    Negative Logits
    matically
    -0.91
    lete
    -0.81
    cking
    -0.79
    mers
    -0.76
    elsen
    -0.74
    ling
    -0.73
    rics
    -0.73
    htaking
    -0.72
    lings
    -0.69
    heit
    -0.69
    POSITIVE LOGITS
    ãģĦ
    1.14
    ãĤī
    1.10
    ãĤĵ
    1.07
    IONS
    1.03
    é»Ĵ
    0.99
    ãģĵ
    0.98
    ãĢĤ
    0.95
    ãģı
    0.92
    ãĤĬ
    0.90
    ãĤ
    0.88
    Activations Density 0.033%

    No Known Activations