© 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. 8-RES_POST_32K-OAI
    11. 6684
    Prev
    Next
    INDEX
    Explanations

    identifiers and numerical values related to code or data structures

    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
    60
    +0.09
    0.4%
    719
    +0.09
    0.4%
    561
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.09
    0.03
    6
    +0.09
    0.04
    488
    +0.08
    0.03
    Negative Logits
    oeuv
    -0.90
    pees
    -0.89
    hops
    -0.85
    abouts
    -0.82
    ogun
    -0.79
    ollah
    -0.75
    enium
    -0.73
    inate
    -0.71
    ilion
    -0.68
    oplan
    -0.68
    POSITIVE LOGITS
     Error
    0.85
     WARN
    0.78
    >]
    0.77
    },{"
    0.73
     error
    0.72
     omitted
    0.72
     Code
    0.70
     mismatch
    0.69
    ]:
    0.67
    Error
    0.67
    Activations Density 0.109%

    No Known Activations