© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4-RES-JB
    11. 20907
    Prev
    Next
    INDEX
    Explanations

    words related to distrust and suspicion in social contexts

    oai_token-act-pair · gpt-3.5-turbo

    themes related to distrust and mistrust in various contexts

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

    No Comments

    Negative Logits
    Redditor
    -0.68
     injunction
    -0.64
    âĹ¼
    -0.64
    education
    -0.62
    ammy
    -0.62
    gran
    -0.61
    amins
    -0.60
    ODE
    -0.59
    Interstitial
    -0.59
    ectar
    -0.58
    POSITIVE LOGITS
    ful
    1.23
    fully
    1.19
    lessly
    1.19
    fulness
    1.08
    lessness
    0.95
    yip
    0.93
    lust
    0.88
    uous
    0.88
    worthiness
    0.86
    chery
    0.82
    Activations Density 0.019%

    No Known Activations