© 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. 3-RES-JB
    11. 12865
    Prev
    Next
    INDEX
    Explanations

    mentions of or related to the word "cone", possibly in the context of shapes or objects

    oai_token-act-pair · gpt-3.5-turbo

    references to cones in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    isons
    -0.88
     Skydragon
    -0.87
    omon
    -0.82
    ties
    -0.78
    aku
    -0.76
    ucks
    -0.74
    anging
    -0.72
    estate
    -0.70
    imaru
    -0.67
    uga
    -0.67
    POSITIVE LOGITS
    cone
    0.76
     cone
    0.72
     Clicker
    0.72
    Spread
    0.70
    REDACTED
    0.66
    OTUS
    0.63
    Sov
    0.62
     Racer
    0.62
    pire
    0.61
    }}}
    0.61
    Activations Density 0.025%

    No Known Activations