© 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. 0-RES-JB
    11. 18916
    Prev
    Next
    INDEX
    Explanations

    terms related to sci-fi and fantasy genres

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the prefix "sci-fi."

    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
     paddle
    -0.71
     oval
    -0.70
     ske
    -0.70
     batters
    -0.69
     squared
    -0.65
     pose
    -0.65
     ranks
    -0.64
     Straw
    -0.64
     batter
    -0.64
     Ging
    -0.63
    POSITIVE LOGITS
    fi
    4.47
    Fi
    2.19
     fi
    1.86
    FI
    1.58
     Fi
    1.48
    ifi
    1.34
    fy
    1.31
    fa
    1.28
    fs
    1.14
    ti
    1.13
    Activations Density 0.008%

    No Known Activations