INDEX

Explanations

repeated characters or patterns within text segments

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

Values with percent signs and numerical ranges

np_acts-logits-general · gemini-2.0-flash

health interventions and outcomes

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

 ComVisible

-0.72

$",

-0.67

Manbalar

-0.66

{}",

-0.65

 يتيمه

-0.64

 estekak

-0.62

————

-0.60

=$((

-0.60

———-

-0.58

 Audiodateien

-0.58

POSITIVE LOGITS

�

0.88

 ��

0.82

 oars

0.68

️

0.65

 Shaksp

0.64

Cze

0.64

 lemmas

0.63

 Arcane

0.63

 gubern

0.61

ValueStyle

0.61

Activations Density 0.060%

repeated characters or patterns within text segments

Values with percent signs and numerical ranges

health interventions and outcomes

No Comments

No Known Activations