INDEX

Explanations

instances of the word "after" in various contexts

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

unda

-0.15

zelf

-0.15

cket

-0.15

rowse

-0.15

lah

-0.14

ifying

-0.14

shiv

-0.14

orning

-0.14

adal

-0.14

nin

-0.13

POSITIVE LOGITS

thought

0.23

wards

0.21

words

0.20

TRA

0.18

word

0.17

/current

0.17

noon

0.16

NO

0.15

reno

0.15

effects

0.15

Activations Density 0.118%

instances of the word "after" in various contexts

No Comments

No Known Activations

instances of the word "after" in various contexts

No Comments

No Known Activations