INDEX

Explanations

statements that assert conclusions or outcomes

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

time

-0.16

 thorough

-0.15

rose

-0.15

ai

-0.15

ÑĬ

-0.14

 Ð²ÐµÐ´ÑĮ

-0.14

tec

-0.14

rot

-0.14

ton

-0.14

Optimizer

-0.14

POSITIVE LOGITS

forth

0.25

ìį¨

0.16

ìĿ¸ì§Ģ

0.15

uger

0.15

fter

0.15

 latter

0.14

gether

0.14

getManager

0.14

odox

0.14

ìĦľ

0.14

Activations Density 0.025%

statements that assert conclusions or outcomes

No Comments

No Known Activations