INDEX

Explanations

words and suffixes related to "tion" and "ent" endings, indicating actions or processes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

olar

-0.18

åĲĪ

-0.15

 åĥ

-0.15

EA

-0.15

alm

-0.14

apel

-0.13

STA

-0.13

ãĥ³ãĥĨãĤ£

-0.13

pls

-0.13

acas

-0.13

POSITIVE LOGITS

naire

0.18

vos

0.15

å·

0.15

WithContext

0.15

gages

0.14

ssp

0.13

ilerine

0.13

Ä±sÄ±na

0.13

jom

0.13

imin

0.13

Activations Density 0.016%

words and suffixes related to "tion" and "ent" endings, indicating actions or processes

No Comments

No Known Activations