INDEX

Explanations

repeated phrases and references to specific examples or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

thon

-0.15

ABCDEFGHI

-0.14

 addCriterion

-0.14

çģµ

-0.14

íĮ

-0.13

	damage

-0.13

(#)

-0.13

RunWith

-0.13

Ã¶l

-0.13

åºŁ

-0.13

POSITIVE LOGITS

 kind

0.38

 kinds

0.33

 type

0.33

kind

0.27

-type

0.26

type

0.25

 types

0.23

 sorts

0.23

 sort

0.23

 exact

0.21

Activations Density 0.189%

repeated phrases and references to specific examples or concepts

No Comments

No Known Activations