© 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. 1-RES-JB
    11. 3907
    Prev
    Next
    INDEX
    Explanations

    text related to the Raspberry Pi, a small single-board computer

    oai_token-act-pair · gpt-3.5-turbo

    references to the Raspberry Pi

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tyard
    -0.94
    lined
    -0.86
    lines
    -0.82
    title
    -0.73
    */(
    -0.72
    ahead
    -0.72
    idas
    -0.71
    Interstitial
    -0.69
    person
    -0.69
    hart
    -0.69
    POSITIVE LOGITS
     Pi
    1.35
    Pi
    1.19
     Pis
    1.09
    ñ
    1.07
    ña
    0.99
     pi
    0.97
    achu
    0.91
    yip
    0.87
    ilon
    0.87
    ÅĤ
    0.85
    Activations Density 0.006%

    No Known Activations