INDEX

Explanations

phrases related to hypothetical situations or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ëĤĺëĬĶ

-0.14

 wordt

-0.13

ombres

-0.13

ffi

-0.13

;++

-0.13

isko

-0.12

Ð·Ð½Ð°

-0.12

ÎºÎ¿Ïį

-0.12

Ð»ÑıÑİÑĤÑģÑı

-0.12

 Affero

-0.12

POSITIVE LOGITS

 have

0.63

’ve

0.54

have

0.53

've

0.53

 telah

0.52

Have

0.48

 Have

0.47

 ÄĳÃ£

0.46

 hebben

0.46

.have

0.43

Activations Density 0.247%

phrases related to hypothetical situations or conditions

No Comments

No Known Activations