INDEX

Explanations

conjunctions and words indicating connection or contrast

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

arih

-0.17

ugu

-0.16

-0.15

ç¹ģ

-0.15

lech

-0.15

 Hess

-0.14

aris

-0.14

illon

-0.14

ØªØ¹

-0.14

mares

-0.14

POSITIVE LOGITS

áº¢

0.15

&&&&

0.15

ress

0.15

 Sabb

0.14

yg

0.14

Ø§ÙĤ

0.14

ertas

0.14

 Permanent

0.14

wyn

0.14

erta

0.13

Activations Density 0.120%

conjunctions and words indicating connection or contrast

No Comments

No Known Activations

conjunctions and words indicating connection or contrast

No Comments

No Known Activations