© 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. 13081
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal opinions and preferences

    oai_token-act-pair · gpt-3.5-turbo

    references to romantic relationships and emotional expressions

    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
    NetMessage
    -0.72
    Ire
    -0.65
    $.
    -0.64
    ".[
    -0.63
     arrang
    -0.63
     mathemat
    -0.63
    agra
    -0.62
    çĶ°
    -0.62
     notor
    -0.61
     blat
    -0.59
    POSITIVE LOGITS
     meanwhile
    0.94
     huh
    0.69
    ictive
    0.63
     SM
    0.60
     abound
    0.58
    ?'
    0.57
    ?:
    0.57
     however
    0.55
     Cause
    0.55
    ?,
    0.55
    Activations Density 1.413%

    No Known Activations