INDEX

Explanations

references to 'first' in context of ranking, priority, or initial positions

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

first followed by common phrase words

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 errMsg

-0.61

 statusCode

-0.60

umbo

-0.60

ährung

-0.59

geddon

-0.59

programme

-0.59

 Dekker

-0.59

cheme

-0.59

qued

-0.58

caling

-0.58

POSITIVE LOGITS

 first

1.41

first

1.25

First

1.16

 First

1.16

 FIRST

1.13

FIRST

1.06

 firſt

1.02

 ﬁrst

0.95

 firft

0.87

 getFirst

0.84

Activations Density 0.065%

references to 'first' in context of ranking, priority, or initial positions

first followed by common phrase words

No Comments

No Known Activations