© 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. 8-RES-JB
    11. 18495
    Prev
    Next
    INDEX
    Explanations

    adjectives related to morality and judgment

    oai_token-act-pair · gpt-3.5-turbo

    terms related to social and ethical judgments about actions and practices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Adjectives related to morality and judgment

    Explanation Uploaded by User
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    onut
    -0.69
    ahs
    -0.60
    zar
    -0.59
    agine
    -0.58
    andon
    -0.58
    aper
    -0.57
     pesky
    -0.57
    guyen
    -0.56
    Boot
    -0.56
    rollers
    -0.56
    POSITIVE LOGITS
     enough
    1.11
    enough
    0.83
     Enough
    0.77
     against
    0.75
    ãĥ¼ãĥĨ
    0.74
    izable
    0.74
    nell
    0.74
    isable
    0.72
     compared
    0.71
    ãĥ¼ãĤ¯
    0.71
    Activations Density 0.289%

    No Known Activations