© 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. 4-RES-JB
    11. 506
    Prev
    Next
    INDEX
    Explanations

    words related to the geographical location of "Sunshine Coast."

    oai_token-act-pair · gpt-3.5-turbo

    references to "Sunshine," often associated with places or positive descriptions

    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
    aph
    -0.78
    scient
    -0.76
    vice
    -0.73
    izabeth
    -0.71
    iple
    -0.70
    historic
    -0.70
    usable
    -0.70
    omination
    -0.69
    cend
    -0.68
    araoh
    -0.68
    POSITIVE LOGITS
     Sunshine
    0.99
    shine
    0.88
     Libre
    0.85
     Scotia
    0.81
     Beach
    0.81
     Sunrise
    0.80
     Isles
    0.78
     Cookies
    0.75
    burst
    0.74
     Springs
    0.72
    Activations Density 0.016%

    No Known Activations