© 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. 2175
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a call for social justice and equality

    oai_token-act-pair · gpt-3.5-turbo

    instances of conditional phrases and qualifiers in discourse

    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
    ery
    -0.70
    edia
    -0.70
    eny
    -0.66
    etting
    -0.62
    agram
    -0.61
    eln
    -0.60
    icky
    -0.58
    qus
    -0.58
    gged
    -0.58
    ech
    -0.58
    POSITIVE LOGITS
     except
    1.53
    except
    1.48
     Including
    1.47
     irrespective
    1.35
     including
    1.35
    including
    1.33
     regardless
    1.29
     imaginable
    1.18
     INCLUD
    1.15
    excluding
    1.09
    Activations Density 0.474%

    No Known Activations