© 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. 2-RES-JB
    11. 17401
    Prev
    Next
    INDEX
    Explanations

    references to the media company "VICE" and related terms

    oai_token-act-pair · gpt-3.5-turbo

    the word "VICE" and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ramid
    -0.86
    anus
    -0.76
    itsch
    -0.74
    aan
    -0.67
    alon
    -0.66
     deduction
    -0.66
    aic
    -0.66
    aah
    -0.64
    enium
    -0.64
     \'
    -0.64
    POSITIVE LOGITS
     VICE
    1.31
    VICE
    1.06
     versa
    0.87
     PRESIDENT
    0.81
    etheless
    0.76
    Mask
    0.76
    MAP
    0.74
    mire
    0.69
    OPLE
    0.69
    TI
    0.68
    Activations Density 0.014%

    No Known Activations