Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 1463
    Prev
    Next
    INDEX
    Explanations

    references to emergency services, specifically the number 911

    oai_token-act-pair · gpt-3.5-turbo

    mentions of emergency service calls, specifically related to "911"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    protein
    -0.79
    wikipedia
    -0.79
    chance
    -0.73
    spring
    -0.69
    egal
    -0.68
    imester
    -0.67
    ophon
    -0.67
    raviolet
    -0.67
    glomer
    -0.66
    rug
    -0.66
    POSITIVE LOGITS
     911
    1.26
     dispatcher
    0.85
    911
    0.80
     caller
    0.77
    gencies
    0.74
    ILLE
    0.70
    ħĭ
    0.67
    999
    0.67
     microphones
    0.66
     calls
    0.66
    Activations Density 0.019%

    No Known Activations