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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4-RES-JB
    6. 21111
    Prev
    Next
    INDEX
    Explanations

    references to "DE" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the abbreviation "DE" followed by numbers

    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
     BaseType
    -0.83
    olulu
    -0.82
    toggle
    -0.80
    ãĤ¤ãĥĪ
    -0.80
    ocene
    -0.78
    ãĤ¹ãĥĪ
    -0.78
    uyomi
    -0.77
    ãĤ±
    -0.76
    oulos
    -0.75
     è£ıè¦ļéĨĴ
    -0.75
    POSITIVE LOGITS
    VICE
    1.22
    FAULT
    1.16
    FIN
    1.08
    ATH
    1.01
    POS
    1.00
    VEL
    0.99
    PUT
    0.93
    GE
    0.90
    FINE
    0.88
    VE
    0.86
    Activations Density 0.011%

    No Known Activations