INDEX

Explanations

various forms of the word "type."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

vern

-0.17

(disposing

-0.17

wm

-0.14

.thumb

-0.14

 Angelo

-0.14

ãĥ¼ãĥ¬

-0.14

erno

-0.14

idot

-0.14

duce

-0.14

POSITIVE LOGITS

ault

0.20

trl

0.17

ep

0.17

_NR

0.16

epy

0.15

vyk

0.15

 Ward

0.15

iage

0.15

enet

0.15

jab

0.15

Activations Density 0.021%

various forms of the word "type."

No Comments

No Known Activations