© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 5507
    Prev
    Next
    INDEX
    Explanations

    names that end with "gren"

    oai_token-act-pair · gpt-3.5-turbo

    recurrent references to a specific individual's surname

    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
     CPI
    -0.76
     Fundamental
    -0.68
    Trend
    -0.65
     Organisation
    -0.62
     Predict
    -0.61
    WD
    -0.61
    OP
    -0.60
    Lead
    -0.59
     failings
    -0.57
     Cout
    -0.57
    POSITIVE LOGITS
    gren
    1.65
    qv
    1.18
    quist
    1.14
    wagen
    1.08
    strom
    1.03
    borg
    1.03
    issance
    1.02
    ascus
    0.98
    nir
    0.97
    swer
    0.95
    Activations Density 0.011%

    No Known Activations