© 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. 2-RES-JB
    11. 1012
    Prev
    Next
    INDEX
    Explanations

    the word "suspects"

    oai_token-act-pair · gpt-4-turbo

    mentions of suspects, particularly those involved in criminal activities

    oai_token-act-pair · gpt-3.5-turbo

    mentions of suspects involved in incidents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ntil
    -0.77
    vironment
    -0.70
    ammy
    -0.67
    tem
    -0.66
    ann
    -0.66
    abeth
    -0.65
    psey
    -0.65
    owan
    -0.64
    cast
    -0.64
    legraph
    -0.64
    POSITIVE LOGITS
     suspects
    1.08
     suspect
    0.93
     mishand
    0.73
     offenders
    0.71
     suspected
    0.67
    istani
    0.67
    pots
    0.66
     infring
    0.65
     perpetrators
    0.65
     CSI
    0.64
    Activations Density 0.008%

    No Known Activations