© 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. 8-RES-JB
    11. 24364
    Prev
    Next
    INDEX
    Explanations

    phrases that introduce or list points or key information

    oai_token-act-pair · gpt-3.5-turbo

    structured points or lists of observations

    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
    urated
    -0.72
    orate
    -0.70
     Aval
    -0.66
    iership
    -0.63
    urous
    -0.62
     Midlands
    -0.62
    oire
    -0.60
     naughty
    -0.60
    orer
    -0.59
    sterdam
    -0.58
    POSITIVE LOGITS
    âĹı
    1.11
    Firstly
    1.07
     Firstly
    1.02
    Reason
    1.00
    âĢ¢
    0.91
    ·
    0.88
    âĸł
    0.88
    %:
    0.88
    âĺħ
    0.84
    âĻ¦
    0.81
    Activations Density 0.231%

    No Known Activations