© 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. 4300
    Prev
    Next
    INDEX
    Explanations

    instances of the word "by" indicating causation or attribution

    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
    678
    +0.13
    0.6%
    745
    +0.10
    0.4%
    373
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.07
    29
    +0.10
    0.06
    678
    +0.10
    0.05
    Negative Logits
    worthy
    -0.76
     audiences
    -0.72
     Chancellor
    -0.72
     Trilogy
    -0.69
     challengers
    -0.67
    imens
    -0.67
     convictions
    -0.66
     whistleblowers
    -0.66
    arta
    -0.64
     Ware
    -0.62
    POSITIVE LOGITS
     subtract
    0.95
     interpol
    0.88
     swapping
    0.85
     inserting
    0.84
     multiplying
    0.83
     adding
    0.83
     loading
    0.82
     rubbing
    0.82
     applying
    0.81
     emitting
    0.79
    Activations Density 0.521%

    No Known Activations