© 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. 4187
    Prev
    Next
    INDEX
    Explanations

    specific measurements of time and distance

    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
    745
    +0.16
    0.7%
    306
    +0.10
    0.4%
    402
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.16
    0.03
    183
    +0.10
    0.02
    545
    +0.10
    0.02
    Negative Logits
     goodbye
    -0.68
    ammy
    -0.63
    lav
    -0.62
    DAY
    -0.60
     uncond
    -0.60
     bearer
    -0.59
     farewell
    -0.57
    ulet
    -0.55
    ãĥĨ
    -0.55
    peat
    -0.55
    POSITIVE LOGITS
     confines
    1.08
     radius
    0.94
     bounds
    0.93
    ombat
    0.84
     vicinity
    0.82
     boundaries
    0.80
     precincts
    0.78
    paces
    0.78
     limits
    0.76
    roach
    0.74
    Activations Density 0.123%

    No Known Activations