© 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. 6929
    Prev
    Next
    INDEX
    Explanations

    the pronoun "it" and its association in various contexts

    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
    361
    +0.10
    0.5%
    330
    +0.09
    0.4%
    610
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.10
    0.03
    567
    +0.09
    0.02
    15
    +0.09
    0.03
    Negative Logits
     Fowler
    -0.68
    rent
    -0.64
     Cosponsors
    -0.62
     muster
    -0.61
    rir
    -0.61
     exhaustion
    -0.59
    itsch
    -0.58
    Dispatch
    -0.58
    IRC
    -0.58
    requires
    -0.56
    POSITIVE LOGITS
     badly
    0.73
    reprene
    0.73
     chance
    0.69
     ASAP
    0.64
     sake
    0.64
     cake
    0.63
     louder
    0.61
     flexibility
    0.60
    chance
    0.60
    luaj
    0.60
    Activations Density 0.447%

    No Known Activations