INDEX

Explanations

instances of the word "when," indicating a focus on time-related phrases or conditions

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

gil

-0.17

agini

-0.16

adaÅŁ

-0.15

olumn

-0.15

dictions

-0.15

ANDOM

-0.14

scape

-0.14

ÐµÑĢÐ¶

-0.14

ussen

-0.14

unately

-0.14

POSITIVE LOGITS

soever

0.25

/if

0.22

abouts

0.20

-Ð»Ð¸Ð±Ð¾

0.19

-Ð½Ð¸Ð±ÑĥÐ´ÑĮ

0.18

ecessary

0.17

ec

0.16

 necessary

0.15

ev

0.15

Ã§

0.15

Activations Density 0.140%

instances of the word "when," indicating a focus on time-related phrases or conditions

No Comments

No Known Activations

instances of the word "when," indicating a focus on time-related phrases or conditions

No Comments

No Known Activations