INDEX

Explanations

recurring mentions of "all" and other inclusive terms

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

ieri

-0.16

acements

-0.16

 certain

-0.15

_firestore

-0.14

stdarg

-0.14

 <!--[

-0.14

erais

-0.14

Certain

-0.13

sst

-0.13

ie

-0.13

POSITIVE LOGITS

dash

0.18

ules

0.17

ä¸ĢåĪĩ

0.16

/all

0.15

ders

0.15

izens

0.14

ltra

0.14

_IOCTL

0.14

utta

0.14

gars

0.13

Activations Density 0.068%

recurring mentions of "all" and other inclusive terms

No Comments

No Known Activations