INDEX

Explanations

occurrences of the word "was."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 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

 houſe

-0.71

 ſtate

-0.71

 autorytatywna

-0.69

 pleaſure

-0.66

 ſche

-0.66

ftance

-0.63

ſelf

-0.62

ſy

-0.61

 ſch

-0.60

 ſtand

-0.60

POSITIVE LOGITS

was

2.84

was

1.67

Was

1.61

Was

1.59

WAS

1.41

 were

1.39

 buvo

1.23

 была

1.12

twas

1.11

 était

1.08

Activations Density 0.305%

occurrences of the word "was."

No Comments

No Known Activations