© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 11315
    Prev
    Next
    INDEX
    Explanations

    phone numbers or attempts to contact someone

    oai_token-act-pair · gpt-3.5-turbo

    references to phone calls and communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adiator
    -0.74
     biome
    -0.72
     Authors
    -0.71
    weights
    -0.71
     competitions
    -0.68
    ulpt
    -0.68
     exhibitions
    -0.68
     divest
    -0.67
     athleticism
    -0.65
     Bench
    -0.65
    POSITIVE LOGITS
     voic
    1.86
     phone
    1.51
     telephone
    1.49
     caller
    1.48
    phone
    1.45
     fax
    1.44
     hotline
    1.41
     Telephone
    1.38
     Caller
    1.33
     dial
    1.30
    Activations Density 0.527%

    No Known Activations