© 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. 14897
    Prev
    Next
    INDEX
    Explanations

    adjectives ending in '-able', '-ary', '-ical', '-less', '-ive', '-ier', '-ist', '-y', '-ous', and '-ed'

    oai_token-act-pair · gpt-3.5-turbo

    adjectives and derivatives indicating characteristics or qualities

    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
    atra
    -0.78
    Patch
    -0.72
    ften
    -0.71
    veland
    -0.70
    arty
    -0.69
    ixels
    -0.68
    å°Ĩ
    -0.67
    DN
    -0.66
    slave
    -0.66
    Rush
    -0.65
    POSITIVE LOGITS
     behav
    1.04
     behaviour
    0.83
     behaviours
    0.81
     propos
    0.76
     destro
    0.74
     endeavour
    0.71
     pse
    0.70
    theless
    0.69
     behavior
    0.69
     epigen
    0.67
    Activations Density 0.213%

    No Known Activations