© 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. 6495
    Prev
    Next
    INDEX
    Explanations

    references to small, subtle, or inconsequential details

    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
    420
    +0.12
    0.5%
    251
    +0.10
    0.4%
    313
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.12
    0.18
    720
    +0.10
    0.14
    546
    +0.10
    0.08
    Negative Logits
    lyn
    -0.79
    ukong
    -0.71
     Gleaming
    -0.70
    ierce
    -0.66
    !:
    -0.65
    ®,
    -0.65
    endon
    -0.64
    "]=>
    -0.64
    ibur
    -0.64
    amia
    -0.63
    POSITIVE LOGITS
     insignificant
    1.32
     negligible
    1.23
     meager
    1.11
     barely
    1.09
     scant
    0.99
     obscurity
    0.99
     hardly
    0.97
     indist
    0.95
     harmless
    0.94
     sparse
    0.93
    Activations Density 3.729%

    No Known Activations