© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 35199
    Prev
    Next
    INDEX
    Explanations

    specific company or financial terms related to economic news

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    554
    +0.06
    0.6%
    326
    +0.05
    0.6%
    688
    +0.05
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.06
    0.16
    11
    +0.05
    0.04
    57
    +0.05
    0.05
    Negative Logits
    hai
    -0.38
    wered
    -0.37
    ios
    -0.35
    ly
    -0.34
    apo
    -0.34
    eks
    -0.33
    anyahu
    -0.33
    eful
    -0.32
    fund
    -0.32
    ities
    -0.31
    POSITIVE LOGITS
    ĸļ
    0.45
    ANCE
    0.44
    LECT
    0.43
    UTION
    0.42
    BOOK
    0.40
    TEXT
    0.39
    BACK
    0.39
    ORED
    0.39
    COL
    0.38
    ISION
    0.38
    Activations Density 4.511%

    No Known Activations