INDEX

Explanations

phrases conveying a sense of conclusion and engagement with the audience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

amil

-0.18

amm

-0.16

OTO

-0.15

-routing

-0.15

otive

-0.15

statt

-0.15

nam

-0.15

ama

-0.15

 Weiter

-0.15

è¯Ŀ

-0.14

POSITIVE LOGITS

COPE

0.15

ä»¶

0.14

etz

0.14

intl

0.13

_ASSUME

0.13

plat

0.13

 sanity

0.13

LineStyle

0.13

 Bray

0.13

ãĤ¤ãĤ¯

0.13

Activations Density 0.099%

phrases conveying a sense of conclusion and engagement with the audience

No Comments

No Known Activations