INDEX

Explanations

second-person pronouns and instructions directed at the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

adem

-0.17

StateException

-0.15

ahat

-0.14

WARD

-0.14

/WebAPI

-0.13

à¥ĭà¤§

-0.13

Î¸Î®

-0.12

elters

-0.12

imore

-0.12

zheimer

-0.12

POSITIVE LOGITS

 should

0.26

 MUST

0.23

 must

0.22

may

0.22

 will

0.22

 cand

0.22

 need

0.20

wil

0.20

should

0.20

 probably

0.20

Activations Density 0.085%

second-person pronouns and instructions directed at the reader

No Comments

No Known Activations