© 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. 9185
    Prev
    Next
    INDEX
    Explanations

    words related to physical attributes or conditions

    oai_token-act-pair · gpt-3.5-turbo

    terms related to physical conditions and experiences

    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
     Tul
    -0.73
    oi
    -0.71
     Prospect
    -0.69
    ries
    -0.69
     Gos
    -0.68
    ered
    -0.68
     Doctrine
    -0.66
     Aval
    -0.66
    ilts
    -0.65
     Gates
    -0.64
    POSITIVE LOGITS
     speaking
    0.82
     assaulted
    0.78
     exting
    0.78
     assaulting
    0.76
     destro
    0.73
     conclud
    0.72
     incapable
    0.72
    anguage
    0.71
     detonated
    0.70
     attracted
    0.70
    Activations Density 0.009%

    No Known Activations