INDEX

Explanations

references to dates and authorship in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

AGR

-0.14

 verw

-0.14

ernet

-0.14

aldi

-0.14

 necessity

-0.14

tains

-0.13

Mog

-0.13

 Mour

-0.13

oct

-0.13

ager

-0.13

POSITIVE LOGITS

ile

0.17

0.16

azo

0.15

@js

0.15

OLF

0.15

plan

0.15

ÏģÎ¿Î½

0.15

Rip

0.14

ptions

0.14

bus

0.14

Activations Density 0.016%

references to dates and authorship in text

No Comments

No Known Activations