INDEX

Explanations

phrases that express comparisons or relationships between concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mmo

-0.15

@(

-0.14

 Sink

-0.14

spiel

-0.14

 ?><?

-0.14

enet

-0.13

amework

-0.13

egis

-0.13

inki

-0.13

âĢĮØ§ÙĨØ¨Ø§Ø±

-0.13

POSITIVE LOGITS

fbe

0.15

 Bullet

0.14

Ã¶n

0.14

ebe

0.14

sic

0.14

ucher

0.14

agle

0.14

ftime

0.14

atel

0.14

fdc

0.13

Activations Density 0.048%

phrases that express comparisons or relationships between concepts

No Comments

No Known Activations