INDEX

Explanations

words related to numerical quantities or values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

slaught

-0.17

erta

-0.17

alus

-0.16

Ð±ÑĢÑı

-0.15

 Stark

-0.15

Ð½Ð°Ð´Ð»ÐµÐ¶

-0.14

ald

-0.14

uki

-0.14

ereo

-0.14

olik

-0.14

POSITIVE LOGITS

YRO

0.18

ted

0.17

azon

0.17

ok

0.16

 framework

0.16

yro

0.16

inem

0.16

Ø¯Ø§

0.15

ast

0.15

framework

0.15

Activations Density 0.018%

words related to numerical quantities or values

No Comments

No Known Activations