© 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. 6483
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "in [number]."

    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
    564
    +0.13
    0.6%
    631
    +0.11
    0.5%
    261
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.10
    362
    +0.11
    0.08
    22
    +0.09
    0.07
    Negative Logits
    uese
    -0.72
     therein
    -0.69
    encies
    -0.67
    Wr
    -0.65
     thereto
    -0.64
    +(
    -0.64
     accordingly
    -0.64
    CDC
    -0.63
     Canaver
    -0.62
    also
    -0.62
    POSITIVE LOGITS
    tained
    0.67
    taining
    0.65
    ãĥĨãĤ£
    0.60
    hower
    0.59
     narrower
    0.57
    initialized
    0.56
     consultation
    0.55
    vol
    0.55
     ornament
    0.54
     Schwar
    0.54
    Activations Density 0.786%

    No Known Activations