© 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. 4-RES-JB
    11. 2455
    Prev
    Next
    INDEX
    Explanations

    specific actions or events happening in news reports or articles

    oai_token-act-pair · gpt-3.5-turbo

    auxiliary verbs and indicators of actions or states

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

    No Comments

    Negative Logits
    )|
    -0.73
     Avenger
    -0.63
    ãĥij
    -0.60
    Reviewer
    -0.58
     nig
    -0.57
     Chevy
    -0.57
    —"
    -0.57
     [
    -0.55
     limp
    -0.55
    Enlarge
    -0.54
    POSITIVE LOGITS
    yip
    0.78
    hement
    0.73
    ategory
    0.67
    laun
    0.66
    bet
    0.64
    senal
    0.62
    illard
    0.61
    isconsin
    0.61
    eret
    0.60
    upuncture
    0.59
    Activations Density 0.570%

    No Known Activations