INDEX

Explanations

forms of the verb "to be"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

DataMember

-0.51

awaiter

-0.50

 themselves

-0.49

 selectAll

-0.48

Pyrene

-0.48

terape

-0.47

RefNanny

-0.47

styleUrls

-0.47

 they

-0.47

 mereka

-0.46

POSITIVE LOGITS

is

0.91

is

0.82

IS

0.72

Is

0.71

Is

0.68

IS

0.65

 nakalista

0.62

nis

0.62

isk

0.61

icks

0.61

Activations Density 1.063%

forms of the verb "to be"

No Comments

No Known Activations