© 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. 7603
    Prev
    Next
    INDEX
    Explanations

    Python code structure, specifically function definitions and initialization methods

    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
    717
    +0.11
    0.5%
    310
    +0.11
    0.5%
    645
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.03
    158
    +0.11
    0.04
    699
    +0.10
    0.03
    Negative Logits
    CVE
    -0.67
     UNCLASSIFIED
    -0.64
     disg
    -0.63
    confirmed
    -0.59
    FIN
    -0.57
     Forestry
    -0.57
     ALSO
    -0.56
    psc
    -0.54
    CTV
    -0.54
     Canal
    -0.54
    POSITIVE LOGITS
    ():
    1.38
    %:
    1.10
    ]:
    0.94
    ":
    0.93
     nutshell
    0.92
    ):
    0.91
    ':
    0.90
     ):
    0.85
    hedon
    0.74
    ħĭ
    0.72
    Activations Density 0.253%

    No Known Activations