© 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. 8-RES-JB
    11. 10846
    Prev
    Next
    INDEX
    Explanations

    instances of hypothetical or speculative statements

    oai_token-act-pair · gpt-3.5-turbo

    conditional statements and inquiries about future actions

    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
    clerosis
    -0.65
    arms
    -0.61
    sharp
    -0.59
    76561
    -0.59
    Sharp
    -0.58
     doubts
    -0.58
    artisan
    -0.58
    zig
    -0.58
    tun
    -0.57
    gin
    -0.57
    POSITIVE LOGITS
     entail
    0.82
     fare
    0.77
     hereafter
    0.74
    dos
    0.73
     realistically
    0.72
    ember
    0.71
     repay
    0.71
     next
    0.70
     react
    0.69
     RTX
    0.69
    Activations Density 0.168%

    No Known Activations