© 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. 14742
    Prev
    Next
    INDEX
    Explanations

    phrases related to protection or defense against various threats or risks

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express protection from various threats or dangers

    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
    gow
    -0.82
     Cosponsors
    -0.80
    wine
    -0.78
    soDeliveryDate
    -0.77
    ULAR
    -0.72
    LV
    -0.70
    holm
    -0.69
    Chel
    -0.67
    itcher
    -0.67
    resy
    -0.66
    POSITIVE LOGITS
     anybody
    0.77
     anyone
    0.73
     overhe
    0.73
     eaves
    0.71
     preventing
    0.67
     discrimination
    0.67
     infer
    0.67
     nil
    0.65
     adjud
    0.64
     recognizing
    0.63
    Activations Density 0.025%

    No Known Activations