© 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. 4-RES-JB
    11. 24184
    Prev
    Next
    INDEX
    Explanations

    locations or directions, specifically related to cardinal and intermediate directions

    oai_token-act-pair · gpt-3.5-turbo

    directional references and regional identifiers

    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
    pload
    -0.81
     surrender
    -0.67
    Happy
    -0.63
    MC
    -0.63
    helps
    -0.62
     Starts
    -0.62
    Sac
    -0.61
    President
    -0.61
    MM
    -0.60
    ioned
    -0.59
    POSITIVE LOGITS
    east
    1.03
    west
    0.97
     Territories
    0.79
    wards
    0.76
     Queensland
    0.75
    osterone
    0.72
    edly
    0.71
    urst
    0.69
    retion
    0.68
     Emirates
    0.68
    Activations Density 0.021%

    No Known Activations