INDEX

Explanations

references to actions or events related to accountability and oversight

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

ils

-0.17

 ourselves

-0.16

our

-0.16

our

-0.15

ds

-0.15

 stuff

-0.15

au

-0.15

 Ð½Ð°ÑĪÐ¸Ñħ

-0.14

il

-0.14

seg

-0.14

POSITIVE LOGITS

 such

0.26

 Such

0.20

Such

0.19

 suche

0.18

 SUCH

0.18

such

0.18

izon

0.16

 bÃ¶yle

0.16

avors

0.16

Tho

0.15

Activations Density 0.303%

references to actions or events related to accountability and oversight

No Comments

No Known Activations