© 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. 7728
    Prev
    Next
    INDEX
    Explanations

    the presence of risks and challenges related to various situations and contexts

    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
    758
    +0.10
    0.4%
    125
    +0.09
    0.4%
    668
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.10
    0.04
    408
    +0.09
    0.04
    578
    +0.09
    0.02
    Negative Logits
    cknowled
    -0.83
     Manufact
    -0.72
    essen
    -0.66
    ithing
    -0.64
    toc
    -0.63
    ciples
    -0.62
     Celebr
    -0.59
     speeches
    -0.59
     tears
    -0.59
     celebrating
    -0.59
    POSITIVE LOGITS
     hazard
    1.08
     threat
    1.08
     danger
    1.02
     risks
    1.00
     risk
    1.00
     peril
    0.91
     dangers
    0.91
    threat
    0.89
     challenge
    0.87
     obstacle
    0.86
    Activations Density 0.335%

    No Known Activations