© 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. 4-RES-JB
    11. 15945
    Prev
    Next
    INDEX
    Explanations

    positive adjectives

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that convey strong, positive, or significant qualities

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

    No Comments

    Negative Logits
    olson
    -0.77
    oise
    -0.77
    ailable
    -0.71
    aver
    -0.71
    udeb
    -0.70
    ext
    -0.69
    avis
    -0.67
    ocratic
    -0.67
    ajor
    -0.67
    decl
    -0.66
    POSITIVE LOGITS
    nesses
    1.06
    NESS
    0.92
    ly
    0.87
     Ways
    0.85
    ness
    0.85
     Hits
    0.83
     Locations
    0.82
     Solutions
    0.80
    LY
    0.79
     Characters
    0.78
    Activations Density 0.328%

    No Known Activations