© 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. 3-RES-JB
    11. 17573
    Prev
    Next
    INDEX
    Explanations

    phrases containing the word "thorny" or related words like "thorn" and "prickly" when discussing various issues or difficulties

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with difficult or problematic issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aternity
    -0.90
    uthor
    -0.86
    ufact
    -0.79
    atern
    -0.79
    ession
    -0.79
    ocese
    -0.77
    iasco
    -0.74
    ults
    -0.73
    ©¶æ
    -0.73
    viron
    -0.72
    POSITIVE LOGITS
     thorn
    1.28
    bats
    0.89
     prick
    0.82
    staking
    0.80
    vine
    0.76
    bush
    0.75
    tar
    0.71
     wedge
    0.71
    iating
    0.71
     pits
    0.70
    Activations Density 0.023%

    No Known Activations