© 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. 7978
    Prev
    Next
    INDEX
    Explanations

    instances of the word "told" and its variations, indicating reported speech or citations

    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
    373
    +0.20
    1.0%
    104
    +0.11
    0.5%
    350
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    257
    +0.20
    0.05
    104
    +0.11
    0.05
    476
    +0.09
    0.04
    Negative Logits
    ij士
    -0.80
    Discussion
    -0.67
    Report
    -0.64
    eus
    -0.63
    raits
    -0.60
    Audio
    -0.59
    gdala
    -0.58
    UGH
    -0.58
    Manager
    -0.58
    ellen
    -0.58
    POSITIVE LOGITS
     SPACE
    0.64
     Deadline
    0.64
    phies
    0.63
     Yon
    0.62
    eworld
    0.61
     DW
    0.61
     Fortune
    0.59
     tomat
    0.59
     ETH
    0.58
     Lift
    0.57
    Activations Density 0.352%

    No Known Activations