© 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. 7-RES-JB
    11. 13434
    Prev
    Next
    INDEX
    Explanations

    phrases or words related to a specific location or entity named "Helvetica" or "Helvetia"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Hel," indicating a focus on a specific topic or entity associated with that term

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Memories
    -0.68
    HAEL
    -0.67
    EED
    -0.65
    æĦ
    -0.64
    EAR
    -0.62
     Cunning
    -0.61
     incidentally
    -0.61
     Eag
    -0.61
    rers
    -0.61
    OUNT
    -0.59
    POSITIVE LOGITS
    pless
    1.01
    mut
    1.01
    ping
    0.96
    mand
    0.95
    ped
    0.91
    strings
    0.90
    ios
    0.88
    idon
    0.88
    ters
    0.86
    tered
    0.84
    Activations Density 0.027%

    No Known Activations