© 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. 0-RES-JB
    11. 22733
    Prev
    Next
    INDEX
    Explanations

    words related to spatial direction (e.g., upward, forward, eastward)

    oai_token-act-pair · gpt-3.5-turbo

    references to directional movement or orientation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sourced
    -0.68
     vape
    -0.63
     Codex
    -0.62
     Ake
    -0.62
     Canon
    -0.61
     estim
    -0.61
     datas
    -0.60
     Prol
    -0.59
     fucking
    -0.57
     Fn
    -0.57
    POSITIVE LOGITS
    ward
    4.73
    wards
    3.55
    WARD
    2.44
    gard
    1.21
    forth
    1.19
    oward
    1.14
    WAR
    1.10
    bound
    1.09
    stairs
    1.03
    chieve
    1.03
    Activations Density 0.021%

    No Known Activations