© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 5045
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences where someone is being photographed or depicted in a certain way

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pose" and its variations

    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
    rawdownloadcloneembedreportprint
    -0.69
     conglomer
    -0.69
    @#&
    -0.62
    Ļ½
    -0.61
    anguages
    -0.61
    mith
    -0.61
     Fiesta
    -0.61
    dll
    -0.60
    ergy
    -0.60
     converter
    -0.58
    POSITIVE LOGITS
    idon
    1.19
     posed
    0.83
    quest
    0.76
    geon
    0.74
     Danger
    0.74
    alis
    0.73
     vulner
    0.70
     dangers
    0.70
     risks
    0.70
     hazards
    0.69
    Activations Density 0.031%

    No Known Activations