INDEX

Explanations

proper nouns, particularly names and titles associated with notable individuals 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 @ 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

prm

-0.15

èĩ£

-0.15

sag

-0.14

otts

-0.14

çļ

-0.14

urst

-0.14

Kon

-0.13

Ë

-0.13

.boot

-0.13

POSITIVE LOGITS

neider

0.19

$MESS

0.18

ìĹ´

0.16

kÄ±r

0.16

egrator

0.15

ENARIO

0.15

DownList

0.15

gli

0.15

ropolis

0.14

-Mart

0.14

Activations Density 0.033%

proper nouns, particularly names and titles associated with notable individuals or concepts

No Comments

No Known Activations