© 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. 3-RES-JB
    11. 10807
    Prev
    Next
    INDEX
    Explanations

    phrases related to control and influence over something/someone

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to control and authority

    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
    ndra
    -0.79
    traumatic
    -0.76
     enthus
    -0.74
    offer
    -0.73
    owler
    -0.73
    osponsors
    -0.71
    dayName
    -0.69
    mpeg
    -0.65
    iatus
    -0.64
     accuse
    -0.63
    POSITIVE LOGITS
     dominion
    0.80
    dom
    0.77
    eer
    0.73
    orship
    0.72
    @@@@
    0.72
    eers
    0.71
     discretion
    0.71
    eering
    0.70
     overseeing
    0.69
    urally
    0.69
    Activations Density 0.234%

    No Known Activations