© 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. 4-RES-JB
    11. 4230
    Prev
    Next
    INDEX
    Explanations

    words related to giving instructions or commands

    oai_token-act-pair · gpt-3.5-turbo

    the repeated phrase structure "I’ll," indicating a focus on self-referential commentary or promises

    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
    senal
    -0.69
     behavi
    -0.67
    lished
    -0.67
     guiActiveUn
    -0.65
    template
    -0.63
    cription
    -0.63
     è£ıç
    -0.63
    Bloomberg
    -0.61
    */(
    -0.59
    quickShipAvailable
    -0.59
    POSITIVE LOGITS
    uminati
    1.18
    oyd
    1.10
    ounge
    1.05
    inois
    0.99
    sburgh
    0.94
    ength
    0.93
    ipop
    0.90
    ibrary
    0.87
    ibr
    0.84
    aday
    0.83
    Activations Density 0.022%

    No Known Activations