INDEX

Explanations

terms related to health risks and medical emergencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

modifiable

-0.16

 Shut

-0.15

_VARIABLE

-0.15

é¼

-0.15

uber

-0.15

ilog

-0.15

ien

-0.15

SizeMode

-0.14

Sox

-0.14

 Fury

-0.14

POSITIVE LOGITS

 Miss

0.18

 missed

0.18

 medicine

0.17

 tell

0.16

 Tell

0.16

 Pearl

0.15

Miss

0.14

endtime

0.14

 miss

0.14

Gee

0.14

Activations Density 0.020%

terms related to health risks and medical emergencies

No Comments

No Known Activations