© 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. 8-RES-JB
    11. 11583
    Prev
    Next
    INDEX
    Explanations

    negative statements expressing what is not expected or desirable

    oai_token-act-pair · gpt-3.5-turbo

    the negation of statements or ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å¥
    -0.76
    éĹĺ
    -0.70
     Tours
    -0.68
     Might
    -0.68
    æ©
    -0.67
    kamp
    -0.66
     Laws
    -0.65
    Writer
    -0.64
    might
    -0.63
     weap
    -0.63
    POSITIVE LOGITS
     necessarily
    1.35
    icable
    1.33
    icably
    1.28
    eworthy
    1.03
    hin
    0.99
     yet
    0.98
    epad
    0.98
    orious
    0.96
     allowed
    0.94
     uncommon
    0.94
    Activations Density 0.149%

    No Known Activations