© 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. 3-RES-JB
    11. 12197
    Prev
    Next
    INDEX
    Explanations

    the word "EVER" written in capital letters

    oai_token-act-pair · gpt-3.5-turbo

    the word "VER" in various contexts

    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
    waukee
    -0.77
    uyomi
    -0.71
    iewicz
    -0.66
    ophys
    -0.66
    antry
    -0.65
    omorph
    -0.65
    EStream
    -0.65
     Alz
    -0.65
    ensing
    -0.63
    aucus
    -0.62
    POSITIVE LOGITS
    VER
    1.42
    theless
    1.01
    VERS
    0.92
    IFIED
    0.85
    IBLE
    0.82
    VO
    0.78
    THING
    0.77
    GO
    0.77
    TAIN
    0.76
    VE
    0.75
    Activations Density 0.008%

    No Known Activations