© 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. 3-RES-JB
    11. 294
    Prev
    Next
    INDEX
    Explanations

    phrases containing specific capitalized words

    oai_token-act-pair · gpt-3.5-turbo

    identifiers or labels associated with various elements or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bilt
    -0.82
    andise
    -0.81
    north
    -0.71
    perty
    -0.70
     Seym
    -0.70
     rare
    -0.67
     biscuits
    -0.66
    ussen
    -0.65
     tyr
    -0.64
     Gamb
    -0.64
    POSITIVE LOGITS
    irect
    1.10
    iots
    1.10
    DEN
    1.06
    ocument
    1.02
    irection
    0.97
    ividual
    0.85
    ID
    0.83
    iotic
    0.83
    ouble
    0.82
    DER
    0.82
    Activations Density 0.010%

    No Known Activations