© 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. 28426
    Prev
    Next
    INDEX
    Explanations

    entities or identifiers related to official documentation or references

    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
    373
    +0.13
    0.6%
    441
    +0.12
    0.5%
    613
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.02
    131
    +0.12
    0.01
    530
    +0.10
    0.01
    Negative Logits
    redients
    -0.93
    idav
    -0.80
    utenberg
    -0.80
    akespeare
    -0.77
    artney
    -0.76
    ichick
    -0.75
    aeper
    -0.74
    nikov
    -0.72
    icion
    -0.71
    iverpool
    -0.71
    POSITIVE LOGITS
    EMOTE
    0.72
    DEBUG
    0.66
     ALP
    0.64
    WARN
    0.62
     Received
    0.60
    Ń
    0.59
     INFO
    0.59
     Wi
    0.58
     SQU
    0.58
    ./
    0.57
    Activations Density 0.055%

    No Known Activations