© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 5907
    Prev
    Next
    INDEX
    Explanations

    references to "chest" 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
    288
    +0.13
    0.6%
    380
    +0.12
    0.5%
    526
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.01
    391
    +0.12
    0.01
    99
    +0.10
    0.01
    Negative Logits
    DEM
    -0.78
     Doodle
    -0.72
    lihood
    -0.69
     CSI
    -0.66
     GOODMAN
    -0.65
     Ake
    -0.63
     Gael
    -0.62
     Hots
    -0.62
     Yel
    -0.62
     warr
    -0.61
    POSITIVE LOGITS
    nut
    1.76
    nuts
    1.64
    erton
    1.13
     cavity
    1.01
    piece
    0.99
    plate
    0.91
    ilage
    0.86
    pieces
    0.82
    mouth
    0.81
    uit
    0.78
    Activations Density 0.033%

    No Known Activations