INDEX

Explanations

instances of the term "false" in various contexts related to misconceptions or misinformation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

adiens

-0.17

shal

-0.16

amar

-0.16

 ActionTypes

-0.16

lp

-0.15

OfClass

-0.15

ram

-0.15

shire

-0.15

zd

-0.15

istles

-0.14

POSITIVE LOGITS

hood

0.27

 positives

0.23

-flag

0.21

 alarms

0.21

-positive

0.20

 alarm

0.19

fully

0.19

/false

0.18

claim

0.17

ivec

0.17

Activations Density 0.032%

instances of the term "false" in various contexts related to misconceptions or misinformation

No Comments

No Known Activations