© 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. 10958
    Prev
    Next
    INDEX
    Explanations

    phrases related to criminal activities or violations of law

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "committed" in relation to various crimes

    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
    hander
    -0.75
     presets
    -0.71
    sama
    -0.69
     complexion
    -0.69
    ellen
    -0.68
    Bey
    -0.67
    reens
    -0.67
    nas
    -0.66
     valves
    -0.66
    anned
    -0.65
    POSITIVE LOGITS
     offences
    0.87
     committed
    0.87
     suicide
    0.86
     adultery
    0.79
     homicide
    0.75
     perpetrated
    0.75
     victim
    0.74
    acts
    0.74
    vict
    0.73
     homicides
    0.71
    Activations Density 0.023%

    No Known Activations