© 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. 12470
    Prev
    Next
    INDEX
    Explanations

    terms related to specific requirements or necessities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "needed."

    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
    imore
    -0.86
    cow
    -0.71
    izon
    -0.69
    eve
    -0.67
    vironment
    -0.67
    vez
    -0.65
    dump
    -0.64
    creen
    -0.64
    bow
    -0.63
    gdala
    -0.63
    POSITIVE LOGITS
    lessly
    0.87
     reinforcements
    0.72
     repairs
    0.70
     assistance
    0.69
     interruption
    0.69
     urgently
    0.67
     unanim
    0.65
     necessities
    0.65
     accommodation
    0.65
     supplies
    0.65
    Activations Density 0.024%

    No Known Activations