INDEX

Explanations

the word "was" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

uarga

-0.61

 Lindley

-0.59

 noix

-0.58

 kanyang

-0.56

 Keluarga

-0.56

 humaine

-0.55

nocześnie

-0.55

 custa

-0.53

 nemico

-0.53

 demeure

-0.53

POSITIVE LOGITS

Was

0.86

Was

0.84

Were

0.76

WAS

0.71

 Were

0.71

SequentialGroup

0.71

Autoritní

0.68

WAS

0.67

 were

0.64

 WERE

0.62

Activations Density 0.140%

the word "was" in various contexts

No Comments

No Known Activations