INDEX

Explanations

the word "led" in various contexts

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

led by

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Hutchinson

-0.53

MessageState

-0.49

 Compañ

-0.49

Snackbar

-0.47

 Mucha

-0.47

 extremely

-0.46

 getItemCount

-0.45

inson

-0.45

 Braunschweig

-0.45

 comfortable

-0.45

POSITIVE LOGITS

Led

1.02

led

0.98

Led

0.97

leads

0.85

Leads

0.83

 lead

0.82

led

0.81

 Lead

0.80

Lead

0.79

 leads

0.78

Activations Density 0.007%

the word "led" in various contexts

led by

No Comments

No Known Activations