INDEX

Explanations

references to various processes involving communication, search, and verification activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ð»ÐµÐº

-0.18

rtl

-0.15

atts

-0.14

uegos

-0.14

inth

-0.14

.mousePosition

-0.14

utters

-0.14

nero

-0.14

chy

-0.14

oland

-0.14

POSITIVE LOGITS

 instead

0.25

instead

0.23

 Instead

0.20

Instead

0.19

 Ð²Ð¼ÐµÑģÑĤ

0.19

pill

0.18

ìĦľëĬĶ

0.17

 purposes

0.15

.gf

0.14

 yerine

0.14

Activations Density 0.241%

references to various processes involving communication, search, and verification activities

No Comments

No Known Activations