INDEX

Explanations

present continuous and current temporal references

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

 previously

-1.16

 originally

-1.11

 initially

-1.05

 ранее

-1.05

 previous

-1.03

 awalnya

-1.02

 then

-1.02

originally

-1.02

previously

-0.98

 sebelumnya

-0.96

POSITIVE LOGITS

weer

0.66

seks

0.59

 gili

0.59

Regret

0.58

værende

0.58

 unrecogn

0.57

varande

0.55

########.

0.54

 oficialmente

0.54

κέ

0.54

Activations Density 0.314%

present continuous and current temporal references

No Comments

No Known Activations