© 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. 2-RES-JB
    11. 3992
    Prev
    Next
    INDEX
    Explanations

    references to terrorist groups like Isis and Boko Haram, names of their leaders, and organizations like Nasa

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific groups or organizations, particularly focused on Islamic extremist groups and NASA

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lishing
    -0.83
    ships
    -0.80
    CAST
    -0.70
     Ballard
    -0.69
    eport
    -0.69
    ez
    -0.67
    lasses
    -0.67
     guiActiveUnfocused
    -0.66
     Scotch
    -0.65
    ship
    -0.64
    POSITIVE LOGITS
     Isis
    0.95
    querque
    0.86
    aeda
    0.86
    qqa
    0.84
     insurgency
    0.79
    aurus
    0.77
    ihad
    0.74
     mosqu
    0.74
    ilon
    0.74
    encia
    0.73
    Activations Density 0.006%

    No Known Activations