© 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. 7240
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions or code snippets involving subtraction or variable assignment

    oai_token-act-pair · gpt-3.5-turbo

    negative signs or indicators in a technical context

    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
    heet
    -0.78
    opher
    -0.77
    pell
    -0.71
     subsequ
    -0.69
    åŃIJ
    -0.67
     defe
    -0.66
    lete
    -0.66
    asketball
    -0.65
    terday
    -0.64
    Magikarp
    -0.64
    POSITIVE LOGITS
    _-
    1.60
    webkit
    1.05
    =-=-=-=-=-=-=-=-
    1.04
    =-=-=-=-
    0.93
     [|
    0.87
    \-
    0.81
    /-
    0.81
    âĢ¢âĢ¢
    0.75
     UID
    0.70
    ISH
    0.70
    Activations Density 0.055%

    No Known Activations