© 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. 4-RES-JB
    11. 2947
    Prev
    Next
    INDEX
    Explanations

    instances of the word "access"

    oai_token-act-pair · gpt-3.5-turbo

    terms related to access or barriers in various contexts

    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
    assetsadobe
    -0.59
     [+]
    -0.59
    WARD
    -0.57
     Lent
    -0.56
    than
    -0.55
    ±
    -0.54
    enegger
    -0.54
     persuasion
    -0.54
    ŃĶ
    -0.52
    Medium
    -0.52
    POSITIVE LOGITS
    ogie
    0.60
     closer
    0.57
     farther
    0.56
    icum
    0.56
     unin
    0.55
     shine
    0.54
    olation
    0.53
     accountable
    0.53
    ibel
    0.53
     away
    0.53
    Activations Density 0.944%

    No Known Activations