INDEX

Explanations

repetitions of the word "fact."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

illac

-0.15

wor

-0.14

XL

-0.14

Injector

-0.14

.habbo

-0.14

EventListener

-0.14

then

-0.13

(strtolower

-0.13

lectron

-0.13

stras

-0.13

POSITIVE LOGITS

 fact

0.30

 Fact

0.21

itious

0.19

Fact

0.19

 that

0.18

izr

0.18

fact

0.18

 bahwa

0.18

uality

0.17

ually

0.17

Activations Density 0.018%

repetitions of the word "fact."

No Comments

No Known Activations