© 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. 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. 1079
    Prev
    Next
    INDEX
    Explanations

    statements involving disapproval or negative judgement

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating dedication and social critique

    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
    anwhile
    -0.59
    swick
    -0.50
    enegger
    -0.47
    udos
    -0.47
    emonium
    -0.46
    allery
    -0.46
    xtap
    -0.46
    ividual
    -0.45
     ensued
    -0.45
     CrossRef
    -0.45
    POSITIVE LOGITS
     idols
    0.51
     wrong
    0.49
     anymore
    0.47
    )]
    0.47
     masturb
    0.45
     raping
    0.45
     beforehand
    0.44
    ?".
    0.44
     stupid
    0.44
     shitty
    0.44
    Activations Density 4.060%

    No Known Activations