INDEX

Explanations

references to years and centuries, particularly in historical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

frag

-0.15

lÄ±

-0.15

def

-0.15

wyn

-0.14

NM

-0.14

tom

-0.14

coming

-0.14

 Numero

-0.14

ãģ¤

-0.14

rows

-0.14

POSITIVE LOGITS

emoc

0.17

 Feed

0.16

æ½ľ

0.15

edList

0.15

rych

0.15

ipeg

0.14

utex

0.14

Ð½ÑıÐ¼Ð¸

0.14

(PORT

0.13

/feed

0.13

Activations Density 0.051%

references to years and centuries, particularly in historical contexts

No Comments

No Known Activations