© 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. 13830
    Prev
    Next
    INDEX
    Explanations

    phrases related to the manner in which actions or situations are approached or handled

    oai_token-act-pair · gpt-3.5-turbo

    expressions describing methods or manners of doing something

    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
    ua
    -0.67
    inately
    -0.65
    rift
    -0.64
    erity
    -0.63
    ibrary
    -0.63
    iture
    -0.63
    igh
    -0.63
    uked
    -0.60
    ined
    -0.60
    pire
    -0.60
    POSITIVE LOGITS
     shape
    0.77
     that
    0.75
    finding
    0.68
    NE
    0.67
     resembling
    0.66
    MAL
    0.64
     manner
    0.64
     reminiscent
    0.64
     Brach
    0.61
    NO
    0.61
    Activations Density 0.064%

    No Known Activations