INDEX

Explanations

references to complex and conflicting emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Ð°ÑĢÐ°ÑĤ

-0.17

avras

-0.15

:async

-0.15

plits

-0.15

ooks

-0.14

ìĪł

-0.14

apl

-0.14

ibbon

-0.14

uye

-0.14

Ã¡tka

-0.14

POSITIVE LOGITS

Ī

0.16

ron

0.15

ICO

0.15

ico

0.14

 relief

0.14

 Nancy

0.14

orge

0.14

 ÙģØ¶

0.14

ym

0.14

iel

0.13

Activations Density 0.392%

references to complex and conflicting emotions

No Comments

No Known Activations