INDEX

Explanations

phrases that indicate warnings or alerts about potential dangers or negative consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

lastic

-0.15

mnt

-0.15

mirror

-0.15

irit

-0.15

ladu

-0.15

.cx

-0.14

à¸±à¸¥

-0.14

iÄĻ

-0.14

irror

-0.13

Gir

-0.13

POSITIVE LOGITS

overrides

0.16

.metro

0.15

aeda

0.15

ople

0.15

ibar

0.15

erm

0.14

Ð¸ÑĤÐµÑĤ

0.14

evenodd

0.14

 Warn

0.14

_REDIRECT

0.14

Activations Density 0.051%

phrases that indicate warnings or alerts about potential dangers or negative consequences

No Comments

No Known Activations