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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11-RES-JB
    6. 15223
    Prev
    Next
    INDEX
    Explanations

    phrases indicating distance or comparison between situations or objects

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing a degree of inadequacy or deficiency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ayne
    -0.77
    hiba
    -0.77
    assian
    -0.75
    alys
    -0.72
    yrs
    -0.71
    KA
    -0.71
     sqor
    -0.70
     Citation
    -0.69
    endez
    -0.68
    queue
    -0.68
    POSITIVE LOGITS
     infall
    0.98
     flawless
    0.98
     conclusive
    0.95
     glamorous
    0.92
     pristine
    0.91
     insignificant
    0.90
     frivolous
    0.90
     conducive
    0.90
     perfect
    0.89
     exhaustive
    0.88
    Activations Density 0.073%

    No Known Activations