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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10-RES-JB
    11. 17319
    Prev
    Next
    INDEX
    Explanations

    terms related to making decisions or calculations based on certain criteria or data

    oai_token-act-pair · gpt-3.5-turbo

    phrases that refer to assessments or conclusions based on prior information or assumptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    zz
    -0.61
    é¾įå
    -0.60
    qv
    -0.58
    osen
    -0.58
    ese
    -0.56
    vision
    -0.55
    TD
    -0.55
     Kro
    -0.55
    ESE
    -0.55
     slit
    -0.55
    POSITIVE LOGITS
     upon
    1.24
     solely
    1.02
     on
    0.95
     loosely
    0.89
     purely
    0.83
    upon
    0.83
     off
    0.83
     squarely
    0.79
    uate
    0.78
     primarily
    0.76
    Activations Density 0.038%

    No Known Activations