INDEX

Explanations

references to television and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ëģĶ

-0.22

 nhau

-0.20

ãĤ©

-0.20

ÌĪ

-0.17

ãĤ§

-0.16

ima

-0.16

ØªÙĩ

-0.16

ert

-0.16

ish

-0.15

ossible

-0.15

POSITIVE LOGITS

mente

0.23

ity

0.19

à¸£

0.18

ately

0.17

naire

0.16

taire

0.16

ally

0.16

aire

0.15

SHIP

0.15

Ð¼Ð¸

0.15

Activations Density 0.674%

references to television and related terms

No Comments

No Known Activations