© 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. 10-RES-JB
    11. 2233
    Prev
    Next
    INDEX
    Explanations

    scientifically-backed evidence or research

    oai_token-act-pair · gpt-3.5-turbo

    references to research and evidence supporting claims

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NetMessage
    -0.70
    antry
    -0.65
    asks
    -0.63
    reddits
    -0.62
    por
    -0.62
    ||||
    -0.62
    fun
    -0.62
     needing
    -0.61
    iband
    -0.60
    isode
    -0.59
    POSITIVE LOGITS
     evidence
    1.74
     studies
    1.70
     empirical
    1.59
     research
    1.47
    evidence
    1.46
    Evidence
    1.41
     Evidence
    1.41
     anecdotal
    1.40
     surveys
    1.33
     study
    1.27
    Activations Density 0.813%

    No Known Activations