© 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. 7-RES-JB
    11. 2988
    Prev
    Next
    INDEX
    Explanations

    people or sources who are familiar with a particular subject or situation

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals who have insider knowledge or familiarity with specific topics or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     1924
    -0.74
     Tooth
    -0.73
     landfall
    -0.71
    tumblr
    -0.71
    Interstitial
    -0.70
    TIT
    -0.67
     mosquitoes
    -0.64
     1931
    -0.64
     billboards
    -0.64
    aceous
    -0.64
    POSITIVE LOGITS
     briefed
    1.05
     priv
    0.84
     familiar
    0.81
     knowledgeable
    0.81
    remlin
    0.79
    arten
    0.76
    ahon
    0.72
    ategy
    0.72
    inion
    0.71
     disclosed
    0.71
    Activations Density 0.141%

    No Known Activations