© 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. 6380
    Prev
    Next
    INDEX
    Explanations

    names associated with specific individuals or entities linked to notable events or topics

    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
    459
    +0.12
    0.6%
    728
    +0.10
    0.5%
    371
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    288
    +0.10
    0.02
    224
    +0.10
    0.01
    Negative Logits
    essee
    -1.00
    arians
    -0.99
    iversity
    -0.93
    ulhu
    -0.91
    ivism
    -0.90
    itude
    -0.88
    ieth
    -0.87
    ions
    -0.87
    arios
    -0.85
    ocular
    -0.84
    POSITIVE LOGITS
    ll
    0.98
    bian
    0.82
    lling
    0.82
    zon
    0.80
    ling
    0.80
    dos
    0.77
    ves
    0.77
    led
    0.75
    tes
    0.74
    lee
    0.74
    Activations Density 0.086%

    No Known Activations