© 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. 11-RES-JB
    11. 14552
    Prev
    Next
    INDEX
    Explanations

    contractions with a negative connotation or prohibiting action

    oai_token-act-pair · gpt-3.5-turbo

    negative contractions indicating denial or refusal

    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
     Morg
    -0.65
     Fatal
    -0.64
     watches
    -0.63
     mathemat
    -0.61
     Shared
    -0.61
     Palest
    -0.60
     Giov
    -0.59
     Investig
    -0.59
     horizont
    -0.59
     Libyan
    -0.57
    POSITIVE LOGITS
    ĸ
    1.29
    £
    1.25
    Ķ
    1.22
    ª
    1.21
    į
    1.20
    Ļ
    1.19
    ©
    1.19
    ĵ
    1.19
    ¬
    1.19
    ¨
    1.17
    Activations Density 0.183%

    No Known Activations