Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 23070
    Prev
    Next
    INDEX
    Explanations

    verbs or nouns related to going through, breaking through, or entering something

    oai_token-act-pair · gpt-3.5-turbo

    terms related to penetrating or penetration

    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
    umble
    -0.75
    inn
    -0.70
    arb
    -0.66
    Lot
    -0.63
    HCR
    -0.62
    quickShipAvailable
    -0.61
    aver
    -0.61
    Hel
    -0.60
    alf
    -0.60
    Ba
    -0.60
    POSITIVE LOGITS
     penetrated
    1.27
     penetration
    1.25
     penetrate
    1.22
     penet
    1.22
     penetrating
    1.11
     Penet
    0.88
     deeper
    0.82
     embr
    0.80
    ipher
    0.78
     piercing
    0.78
    Activations Density 0.017%

    No Known Activations