© 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. 9458
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     pupil
    -0.66
     Cambridge
    -0.64
    "><
    -0.63
     Mr
    -0.63
     ment
    -0.62
     concess
    -0.61
     Moy
    -0.61
     patron
    -0.61
     Canaveral
    -0.59
     à¤
    -0.59
    POSITIVE LOGITS
    ãĥ³ãĤ¸
    0.83
    ãĤ´ãĥ³
    0.74
    agons
    0.74
    Reviewed
    0.73
    agonists
    0.72
     STATS
    0.72
    adish
    0.68
     Takeru
    0.68
    ãĥ¼ãĥĨ
    0.68
    immune
    0.67
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.