© 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. 5022
    Prev
    Next
    INDEX
    Explanations

    forms of the pronoun "I" and its associated verb structures

    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
    509
    +0.12
    0.6%
    314
    +0.09
    0.4%
    516
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    319
    +0.12
    0.04
    155
    +0.09
    0.03
    9
    +0.09
    0.03
    Negative Logits
    undrum
    -0.68
     Photographer
    -0.65
     Dwarf
    -0.65
    aldo
    -0.64
     Reach
    -0.63
     Kod
    -0.61
    IRO
    -0.60
    Loop
    -0.60
     Explorer
    -0.59
    urion
    -0.58
    POSITIVE LOGITS
    rolet
    0.72
    psych
    0.71
     Thus
    0.70
    ec
    0.69
    say
    0.68
     insofar
    0.68
     namely
    0.68
    ologically
    0.67
    acia
    0.67
    Thus
    0.66
    Activations Density 0.278%

    No Known Activations