INDEX

Explanations

phrases indicating a personal opinion or assertion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gren

-0.18

cred

-0.16

vala

-0.15

vid

-0.14

Orb

-0.14

odes

-0.14

Ø§ÙĪØª

-0.14

gmt

-0.14

acea

-0.13

 Ð±Ð°Ð³

-0.13

POSITIVE LOGITS

tout

0.16

sca

0.15

eÅŁ

0.15

ÐºÐ¾ÑĤ

0.15

istÃ©

0.14

(SC

0.14

atori

0.14

toa

0.14

.paging

0.14

onian

0.14

Activations Density 0.156%

phrases indicating a personal opinion or assertion

No Comments

No Known Activations