© 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. 0-RES-JB
    11. 2804
    Prev
    Next
    INDEX
    Explanations

    mentions of criminal activities involving juveniles

    oai_token-act-pair · gpt-3.5-turbo

    mentions of juveniles or references to secrecy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Crusher
    -0.71
    amel
    -0.71
     consolation
    -0.70
     avail
    -0.69
    ={
    -0.69
    ajor
    -0.67
     cele
    -0.66
     spr
    -0.65
    =]
    -0.65
     Peaks
    -0.65
    POSITIVE LOGITS
     juvenile
    1.64
     secretive
    1.44
     archaic
    1.43
     bloated
    1.42
     outdated
    1.32
     intrusive
    1.24
     immature
    1.23
     predatory
    1.22
    ambling
    1.18
     unfocused
    1.17
    Activations Density 0.074%

    No Known Activations