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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4945
    Prev
    Next
    INDEX
    Explanations

    mentions or prompts to sign up for newsletters, clubs, or services

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Sign" followed by variations indicating a call to action

    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
     adventure
    -0.73
     duty
    -0.69
     gul
    -0.68
     humour
    -0.67
     af
    -0.66
     amusement
    -0.66
     aside
    -0.66
     suppose
    -0.65
     barb
    -0.64
     dungeon
    -0.64
    POSITIVE LOGITS
    Sign
    3.96
    sign
    2.40
     Sign
    2.33
    SIGN
    1.99
     SIGN
    1.87
     Signs
    1.84
     sign
    1.65
    signed
    1.47
     signs
    1.40
     Signature
    1.23
    Activations Density 0.015%

    No Known Activations