© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 13066
    Prev
    Next
    INDEX
    Explanations

    verbs related to stopping or ending something

    oai_token-act-pair · gpt-3.5-turbo

    terms related to control or suppression

    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
     amen
    -0.67
    hiba
    -0.66
    ceivable
    -0.63
    inav
    -0.62
    anuts
    -0.62
    ever
    -0.62
    ortment
    -0.60
    uka
    -0.60
    ocese
    -0.59
    Ranked
    -0.57
    POSITIVE LOGITS
     havoc
    0.83
    ieri
    0.77
     weeds
    0.74
     Flavoring
    0.70
    psy
    0.70
    flower
    0.69
    cens
    0.68
     Katrina
    0.66
    balance
    0.65
     tears
    0.63
    Activations Density 0.067%

    No Known Activations