© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 12425
    Prev
    Next
    INDEX
    Explanations

    words related to attacks or aggression

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Att" or variations of it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     shorth
    -0.76
     dating
    -0.62
     timelines
    -0.60
     internally
    -0.58
    SHIP
    -0.58
    terday
    -0.57
     Bosnia
    -0.57
    manship
    -0.55
     POLITICO
    -0.55
     trough
    -0.55
    POSITIVE LOGITS
    orneys
    1.61
    ractive
    1.59
    itudinal
    1.52
    itudes
    1.52
    raction
    1.48
    achment
    1.47
    ention
    1.44
    itude
    1.44
    itud
    1.37
    ractions
    1.36
    Activations Density 0.026%

    No Known Activations