© 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. 17132
    Prev
    Next
    INDEX
    Explanations

    words related to attempting or testing

    oai_token-act-pair · gpt-3.5-turbo

    instances of attempts or invitations to experience or sample something new

    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
    Jews
    -0.73
    head
    -0.73
    front
    -0.71
    lance
    -0.70
    Newsletter
    -0.69
    fore
    -0.69
     Debor
    -0.68
    lot
    -0.63
    scribe
    -0.63
    cens
    -0.63
    POSITIVE LOGITS
     unsuccessfully
    0.95
    avorite
    0.78
    amins
    0.77
     experimenting
    0.75
    nir
    0.74
     refreshing
    0.70
    76561
    0.70
     carbohyd
    0.69
    ãĥīãĥ©
    0.67
    bles
    0.67
    Activations Density 0.049%

    No Known Activations