INDEX

Explanations

phrases and words indicating size, particularly small or brief items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

.setViewport

-0.16

isa

-0.15

UGHT

-0.15

acher

-0.15

à¦°

-0.15

tek

-0.15

rub

-0.15

osu

-0.14

acia

-0.14

POSITIVE LOGITS

(<

0.21

ãģªãģĮãĤī

0.19

weg

0.17

(~

0.15

_COMPAT

0.15

ç¨ĭåº¦

0.15

 thÃ´i

0.14

edReader

0.14

avery

0.14

appa

0.14

Activations Density 0.145%

phrases and words indicating size, particularly small or brief items

No Comments

No Known Activations