© 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. 3388
    Prev
    Next
    INDEX
    Explanations

    phrases related to suspicion or suspicion of criminal activity

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to individuals being accused or believed to be involved in criminal activities

    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
    learn
    -0.78
    idge
    -0.76
    lite
    -0.74
    reen
    -0.73
    ajo
    -0.73
    rx
    -0.71
    ammy
    -0.71
    mar
    -0.71
    alach
    -0.69
    idity
    -0.66
    POSITIVE LOGITS
     suspect
    0.87
     suspects
    0.81
     culprit
    0.76
     mishand
    0.70
     suspected
    0.69
     accuse
    0.68
     offenders
    0.67
     accused
    0.66
     misuse
    0.66
     wrongly
    0.65
    Activations Density 0.016%

    No Known Activations