© 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. 7915
    Prev
    Next
    INDEX
    Explanations

    significant events or actions involving violence and conflict

    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
    629
    +0.12
    0.5%
    65
    +0.10
    0.5%
    541
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.12
    0.01
    211
    +0.10
    0.01
    373
    +0.10
    -0.01
    Negative Logits
    Pg
    -0.76
    cephal
    -0.72
     pse
    -0.72
    Pand
    -0.61
    ROM
    -0.61
     bitters
    -0.61
     bitterly
    -0.59
     navig
    -0.59
    hett
    -0.59
     proxies
    -0.58
    POSITIVE LOGITS
     explosives
    0.81
     è£ıè
    0.79
    APD
    0.78
     ATF
    0.72
     Explosion
    0.68
    stasy
    0.68
     Investigators
    0.67
    angan
    0.67
     SUV
    0.65
     explosion
    0.65
    Activations Density 0.042%

    No Known Activations