© 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. 7148
    Prev
    Next
    INDEX
    Explanations

    terms related to instigating conflict or tension

    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
    288
    +0.13
    0.6%
    247
    +0.12
    0.5%
    730
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.05
    271
    +0.12
    0.03
    247
    +0.12
    0.03
    Negative Logits
    Leaks
    -0.87
    baum
    -0.71
    prints
    -0.71
    leaf
    -0.70
    cise
    -0.69
    oard
    -0.69
    aum
    -0.69
     Accounting
    -0.68
    eem
    -0.68
    éĸ
    -0.68
    POSITIVE LOGITS
     retaliation
    1.01
     escalation
    0.99
     aggression
    0.94
     escalate
    0.89
     escal
    0.89
     aggress
    0.88
     provocation
    0.88
     provoked
    0.87
     confrontation
    0.87
     bellig
    0.87
    Activations Density 0.210%

    No Known Activations