INDEX

Explanations

names, particularly those associated with individuals in various contexts, including scientific literature

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

creativecommons

-0.15

egg

-0.15

veau

-0.14

oard

-0.14

erville

-0.14

etin

-0.14

ertino

-0.14

Î´Î¯

-0.14

andin

-0.14

.cz

-0.13

POSITIVE LOGITS

dbl

0.16

IIIK

0.14

éĥİ

0.14

lit

0.14

mdi

0.14

ÐµÐ²Ð¸Ñĩ

0.13

_mk

0.13

 pseud

0.13

_AUTO

0.13

Activations Density 0.121%

names, particularly those associated with individuals in various contexts, including scientific literature

No Comments

No Known Activations