© 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. 9146
    Prev
    Next
    INDEX
    Explanations

    phrases related to expressing or determining one's position on a topic or issue

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to political positions or beliefs

    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
    teness
    -0.92
    avage
    -0.65
    osi
    -0.63
    voy
    -0.62
    mens
    -0.61
     mention
    -0.60
     mayors
    -0.60
    endor
    -0.60
    iability
    -0.59
    signed
    -0.57
    POSITIVE LOGITS
     Corm
    0.68
     depends
    0.67
    áµ
    0.64
    alian
    0.62
    awei
    0.61
     unfolded
    0.61
    ratom
    0.59
     differs
    0.59
     proced
    0.58
     unfolding
    0.58
    Activations Density 0.397%

    No Known Activations