INDEX

Explanations

variations of the word "Val" or similar patterns, possibly related to names or labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

elly

-0.19

eling

-0.19

eting

-0.18

elli

-0.17

elson

-0.17

esse

-0.16

idon

-0.16

elle

-0.16

ese

-0.15

eca

-0.15

POSITIVE LOGITS

entine

0.28

entina

0.26

uation

0.24

entin

0.23

leys

0.23

uable

0.23

uations

0.21

val

0.21

ued

0.20

=val

0.20

Activations Density 0.025%

variations of the word "Val" or similar patterns, possibly related to names or labels

No Comments

No Known Activations