INDEX

Explanations

dates and years

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

irie

-0.16

ãĥ³ãĥĶ

-0.15

getVar

-0.14

oplan

-0.14

ÏĥÏī

-0.14

 Wars

-0.13

ened

-0.13

ĸ

-0.13

Flo

-0.13

apat

-0.13

POSITIVE LOGITS

ument

0.16

lc

0.16

statt

0.15

gress

0.15

GRESS

0.15

Î¬Î»Î¹

0.14

Ð¼ÐµÐ¶

0.14

Č

0.14

_HIDE

0.14

RESS

0.14

Activations Density 0.075%

dates and years

No Comments

No Known Activations