INDEX

Explanations

expressions of strong enthusiasm or emphasis

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

ese

-0.19

 mere

-0.16

ile

-0.16

strand

-0.15

 Dear

-0.15

osa

-0.14

chwitz

-0.14

ension

-0.14

elight

-0.14

aid

-0.14

POSITIVE LOGITS

ignment

0.21

ationship

0.20

igned

0.17

-life

0.17

tek

0.17

-world

0.16

-ÑĤÐ°ÐºÐ¸

0.16

 estate

0.16

uger

0.15

uste

0.15

Activations Density 0.049%

expressions of strong enthusiasm or emphasis

No Comments

No Known Activations