© 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. 3942
    Prev
    Next
    INDEX
    Explanations

    phrases related to things or people coming under scrutiny or criticism

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating situations of scrutiny or criticism

    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
    grade
    -0.62
    ga
    -0.55
    chat
    -0.55
    IGN
    -0.54
    ashington
    -0.54
    atto
    -0.54
    QL
    -0.54
     Tale
    -0.54
    geries
    -0.53
    ifted
    -0.52
    POSITIVE LOGITS
    unal
    0.74
    BuyableInstoreAndOnline
    0.67
    lehem
    0.64
    quished
    0.63
    å§
    0.60
    noticed
    0.60
    Interstitial
    0.59
    EVA
    0.59
    éģ
    0.58
    ð
    0.58
    Activations Density 0.134%

    No Known Activations