INDEX

Explanations

terms related to measurement and categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

Æ°á»Ľi

-0.14

cery

-0.14

ufe

-0.13

.TestTools

-0.13

_HIDE

-0.13

ebo

-0.12

uchar

-0.12

WP

-0.12

UTH

-0.12

lon

-0.12

POSITIVE LOGITS

#aa

0.16

portun

0.16

Ã²a

0.16

#ae

0.15

hei

0.15

FromArray

0.15

Vie

0.15

scribe

0.14

#af

0.14

rlen

0.14

Activations Density 0.044%

terms related to measurement and categorization

No Comments

No Known Activations