© 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. 10-RES_POST_32K-OAI
    11. 28560
    Prev
    Next
    INDEX
    Explanations

    words related to benefits and positive impacts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    149
    +0.13
    0.6%
    555
    +0.13
    0.6%
    97
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    555
    +0.13
    0.03
    314
    +0.13
    -0.00
    288
    +0.10
    0.03
    Negative Logits
    é¾
    -0.87
     aside
    -0.74
     plural
    -0.66
     Catal
    -0.63
     Hebdo
    -0.62
     Flat
    -0.62
     Falk
    -0.62
     rall
    -0.60
     Logan
    -0.60
     plainly
    -0.60
    POSITIVE LOGITS
    actor
    1.44
    icial
    1.35
    iciary
    1.25
    act
    1.19
    itting
    1.16
    icient
    1.15
    riend
    1.13
    itted
    1.11
    ield
    1.10
    iciency
    1.01
    Activations Density 0.174%

    No Known Activations