INDEX

Explanations

phrases related to unusual or sensational news stories

oai_token-act-pair · gpt-4o-mini Triggered by @bot

that introduce surprising or dubious events

np_acts-logits-general · gemini-2.0-flash

news reports about crime or unusual events

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-0.97

findpost

-0.69

psack

-0.66

seamnă

-0.64

TacToe

-0.63

 AssemblyCulture

-0.62

 vertrou

-0.62

HostException

-0.62

 segera

-0.61

ContentAlignment

-0.60

POSITIVE LOGITS

 Guinness

0.96

 bizarre

0.95

 strangest

0.88

 reportedly

0.82

 weird

0.78

 apparently

0.78

 astonishing

0.76

 allegedly

0.75

 unbelievable

0.73

 unusual

0.72

Activations Density 0.329%

phrases related to unusual or sensational news stories

that introduce surprising or dubious events

news reports about crime or unusual events

No Comments

No Known Activations