© 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. 0-RES-JB
    11. 19318
    Prev
    Next
    INDEX
    Explanations

    instances of the term "capable"

    oai_token-act-pair · gpt-3.5-turbo

    references to capability or competence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cheong
    -0.68
    idays
    -0.66
     Old
    -0.64
    aways
    -0.63
    wal
    -0.62
    ieu
    -0.61
     Funk
    -0.60
    church
    -0.60
    politics
    -0.59
    nen
    -0.58
    POSITIVE LOGITS
     capable
    3.41
     incapable
    2.01
     competent
    1.77
     worthy
    1.60
     deserving
    1.53
     able
    1.50
     suitable
    1.42
     equipped
    1.41
     willing
    1.38
     compatible
    1.36
    Activations Density 0.010%

    No Known Activations