INDEX

Explanations

emotional punctuation and expressions of frustration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Yourself

-0.16

ãĤ´ãĥª

-0.15

ient

-0.14

ano

-0.14

_WARNINGS

-0.14

 Fist

-0.14

xed

-0.14

orted

-0.13

exceptions

-0.13

aza

-0.13

POSITIVE LOGITS

ARG

0.35

ARG

0.29

arr

0.28

 sigh

0.27

Ack

0.26

arg

0.26

ugh

0.26

Bah

0.26

ARR

0.25

arg

0.25

Activations Density 0.157%

emotional punctuation and expressions of frustration

No Comments

No Known Activations