© 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. 0-RES-JB
    11. 15081
    Prev
    Next
    INDEX
    Explanations

    terms related to societal or market trends

    oai_token-act-pair · gpt-3.5-turbo

    references to trends

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     deed
    -0.64
    ride
    -0.64
    iam
    -0.64
    icum
    -0.63
     invocation
    -0.62
    ractor
    -0.61
    nn
    -0.60
    partial
    -0.60
     blessing
    -0.60
    arium
    -0.58
    POSITIVE LOGITS
     trends
    3.52
     Trends
    2.32
     patterns
    1.89
     tendencies
    1.62
     developments
    1.57
     Patterns
    1.56
     demographics
    1.53
     trend
    1.45
    Trend
    1.41
     habits
    1.38
    Activations Density 0.023%

    No Known Activations