© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 6013
    Prev
    Next
    INDEX
    Explanations

    mentions of negative behaviors and practices, especially relating to public perception

    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
    470
    +0.11
    0.5%
    100
    +0.10
    0.4%
    338
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.15
    575
    +0.10
    0.13
    225
    +0.09
    0.10
    Negative Logits
    verning
    -0.84
    imum
    -0.81
    sonian
    -0.77
    negie
    -0.75
    essential
    -0.75
     marvel
    -0.74
    ortium
    -0.73
    idential
    -0.72
    sufficient
    -0.71
     seamless
    -0.71
    POSITIVE LOGITS
     Syndrome
    1.03
     syndrome
    1.00
     perpetrated
    0.97
     plag
    0.77
     inflicted
    0.77
     imaginable
    0.77
     harming
    0.76
     Practices
    0.74
     ruining
    0.74
     plague
    0.74
    Activations Density 2.008%

    No Known Activations